Hi everybody, in http://thread.gmane.org/gmane.emacs.orgmode/47043 I've posted two examples of local file definitions at the top of an org file which are silently ignored for seemingly unrelated parts in the org file. The definitions work if they are put at the end of the org file. This inconsistency is a bug, isn't it? Also, are local variables documented somewhere? I couldn't find anything in the manual. Cheers, Viktor
Never mind, Nick Dokos explained the behavior here: http://thread.gmane.org/gmane.emacs.orgmode/47043/focus=47091 Cheers, Viktor Viktor Rosenfeld wrote: > Hi everybody, > > in http://thread.gmane.org/gmane.emacs.orgmode/47043 I've posted two > examples of local file definitions at the top of an org file which are > silently ignored for seemingly unrelated parts in the org file. The > definitions work if they are put at the end of the org file. > > This inconsistency is a bug, isn't it? Also, are local variables > documented somewhere? I couldn't find anything in the manual. > > Cheers, > Viktor
Viktor Rosenfeld <listuser36@googlemail.com> wrote:
> Hi everybody,
>
> in http://thread.gmane.org/gmane.emacs.orgmode/47043 I've posted two
> examples of local file definitions at the top of an org file which are
> silently ignored for seemingly unrelated parts in the org file. The
> definitions work if they are put at the end of the org file.
>
> This inconsistency is a bug, isn't it? Also, are local variables
> documented somewhere? I couldn't find anything in the manual.
>
No, it's not. Did you see my reply to Darlan's mail last night?
Nick
Nick Dokos <nicholas.dokos@hp.com> wrote:
> Viktor Rosenfeld <listuser36@googlemail.com> wrote:
>
> > Hi everybody,
> >
> > in http://thread.gmane.org/gmane.emacs.orgmode/47043 I've posted two
> > examples of local file definitions at the top of an org file which are
> > silently ignored for seemingly unrelated parts in the org file. The
> > definitions work if they are put at the end of the org file.
> >
> > This inconsistency is a bug, isn't it? Also, are local variables
> > documented somewhere? I couldn't find anything in the manual.
> >
>
> No, it's not. Did you see my reply to Darlan's mail last night?
>
Sorry, I jumped the gun: I should have read further in the list.
Nick
#+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700 * Show code of stored procedure Despite telling Babel that I'd like to see the output as it is, it displays it in a 4-column table. See http://i.imgur.com/neDO3.png for the original layout in SQL Query Analyser (1 column, 34 lines). Babel seems to interpret every *leading space* as *one empty column*. Normal, feature, bug? Is there some workaround to this? I thought stating "scalar" would really completely override any interpretation... ** Code This is an example of code run against a Microsoft SQL server. #+begin_src sql :eval yes :results output scalar EXEC sp_helptext 'dt_setpropertybyid' #+end_src #+results: | Text | | | | |-----------------------------------------+-------------------------------------------------------------------------+---------------------------------------------------------------------------------------------+--------------------------------------------| | /* | | | | | | | | | | ** | If the property already exists, reset the value; otherwise add property | | | | | | | | | ** | | id -- the id in sysobjects of the object | | | | | | | | ** | | property -- the name of the property | | | | | | | | ** | | value -- the text value of the property | | | | | | | | ** | | lvalue -- the binary value of the property (image) | | | | | | | | */ | | | | | | | | | | create procedure dbo.dt_setpropertybyid | | | | | | | | | | | @id int, | | | | | | | | | | @property varchar(64), | | | | | | | | | | @value varchar(255), | | | | | | | | | | @lvalue image | | | | | | | | | as | | | | | | | | | | | set nocount on | | | | | | | | | | declare @uvalue nvarchar(255) | | | | | | | | | | set @uvalue = convert(nvarchar(255), @value) | | | | | | | | | | if exists (select * from dbo.dtproperties | | | | | | | | | | | | where objectid=@id and property=@property) | | | | | | | | begin | | | | | | | | | | | -- | | | | | | | | | | -- bump the version count for this row as we update it | | | | | | | | | | -- | | | | | | | | | | update dbo.dtproperties set value=@value, uvalue=@uvalue, lvalue=@lvalue, version=version+1 | | | | | | | | | | | where objectid=@id and property=@property | | | | | | | | end | | | | | | | | | | else | | | ... Best regards, Seb -- Sebastien Vauban
Hi Seb, I've just pushed up a fix which should resolve this issue. Best -- Eric "Sebastien Vauban" <wxhgmqzgwmuf@spammotel.com> writes: > #+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700 > > * Show code of stored procedure > > Despite telling Babel that I'd like to see the output as it is, it displays it > in a 4-column table. > > See http://i.imgur.com/neDO3.png for the original layout in SQL Query Analyser > (1 column, 34 lines). > > Babel seems to interpret every *leading space* as *one empty column*. Normal, > feature, bug? > > Is there some workaround to this? I thought stating "scalar" would really > completely override any interpretation... > > ** Code > > This is an example of code run against a Microsoft SQL server. > > #+begin_src sql :eval yes :results output scalar > EXEC sp_helptext 'dt_setpropertybyid' > #+end_src > > #+results: > | Text | | | | > |-----------------------------------------+-------------------------------------------------------------------------+---------------------------------------------------------------------------------------------+--------------------------------------------| > | /* | | | | > | | | | | > | ** | If the property already exists, reset the value; otherwise add property | | | > | | | | | > | ** | | id -- the id in sysobjects of the object | | > | | | | | > | ** | | property -- the name of the property | | > | | | | | > | ** | | value -- the text value of the property | | > | | | | | > | ** | | lvalue -- the binary value of the property (image) | | > | | | | | > | */ | | | | > | | | | | > | create procedure dbo.dt_setpropertybyid | | | | > | | | | | > | | @id int, | | | > | | | | | > | | @property varchar(64), | | | > | | | | | > | | @value varchar(255), | | | > | | | | | > | | @lvalue image | | | > | | | | | > | as | | | | > | | | | | > | | set nocount on | | | > | | | | | > | | declare @uvalue nvarchar(255) | | | > | | | | | > | | set @uvalue = convert(nvarchar(255), @value) | | | > | | | | | > | | if exists (select * from dbo.dtproperties | | | > | | | | | > | | | | where objectid=@id and property=@property) | > | | | | | > | | begin | | | > | | | | | > | | | -- | | > | | | | | > | | | -- bump the version count for this row as we update it | | > | | | | | > | | | -- | | > | | | | | > | | | update dbo.dtproperties set value=@value, uvalue=@uvalue, lvalue=@lvalue, version=version+1 | | > | | | | | > | | | | where objectid=@id and property=@property | > | | | | | > | | end | | | > | | | | | > | | else | | | > ... > > Best regards, > Seb -- Eric Schulte http://cs.unm.edu/~eschulte/
Hi Eric,
Eric Schulte wrote:
>> Babel seems to interpret every *leading space* as *one empty column*.
>> Normal, feature, bug?
>>
>> Is there some workaround to this? I thought stating "scalar" would really
>> completely override any interpretation...
>
> I've just pushed up a fix which should resolve this issue.
It does better things, but at least at the wrong place.
--8<---------------cut here---------------start------------->8---
#+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700 :results output
#+begin_src sql :eval yes :results scalar
EXEC sp_helptext 'reset_me'
#+end_src
--8<---------------cut here---------------end--------------->8---
becomes, after evaluation (indented for the sake of clarity):
--8<---------------cut here---------------start------------->8---
#+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700 :results output
Text
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
CREATE PROCEDURE reset_me
@pfi varchar(16)
AS
BEGIN
PRINT 'Done!'
END
#+begin_src sql :eval yes :results scalar
EXEC sp_helptext 'reset_me'
#+end_src
#+results:
#+begin_example
#+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700 :results output
Text
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
CREATE PROCEDURE reset_me
@pfi varchar(16)
AS
BEGIN
PRINT 'Done!'
END
#+begin_src sql :eval yes :results scalar
EXEC sp_helptext 'reset_me'
#+end_src
#+end_example
--8<---------------cut here---------------end--------------->8---
In short, it seems that:
- results is inserted at point's position (in the above case, I was using the
eval speed command `e', thus being at #).
- the results is duplicated, with some part of the source buffer being
repeated as well... See the presence of 2 `#+BABEL:' lines in the buffer,
after the evaluation.
- depending on the point from which I run the code evaluation, there is (or
there isn't) an `#+begin_example' directive. The `#+end_example' is always
present.
Best regards,
Seb
--
Sebastien Vauban
[-- Attachment #1: Type: text/plain, Size: 911 bytes --] Hi, is it possible to export an org file from the command line, so that a currently running Emacs instance is not disturbed? I want to export the attached org file and run the included source blocks, so I have an activity report in the end. I use the shell script pasted below, but there are two problems: - my Emacs instance is blocked during the execution of the shell scripts contained in the file - the script globally sets org-confirm-babel-evaluate to nil for my Emacs instance The second problem could possibly be solved with a local file variable. But the first problem remains. If I use emacs instead of emacsclient, it complains about a running Emacs instance. #!/bin/sh emacsclient -c \ --eval "(progn (find-file \"macports.org\")) (setq org-confirm-babel-evaluate nil) (org-export-as-html 3) (kill-buffer) (delete-frame))" I'm using Org-Mode 7.7. Cheers, Viktor [-- Attachment #2: macports.org --] [-- Type: text/plain, Size: 1200 bytes --] #+TITLE: MacPorts Update Report * Helper functions :noexport: ** Set up =$PATH= environment Set up =$PATH=: #+begin_src sh :session port-update :results silent PATH=$PATH:/opt/local/bin #+end_src ** Table Header for the list of outdated ports #+begin_src sh :session port-update :results silent print_outdated_header() { echo \|Port\|Installed\|Available\| echo "|--|" } #+end_src ** Format each row of =port outdated= as table entry #+begin_src sh :session port-update :results silent format_port_outdated() { awk '{print "|" $1 "|" $2 "|" $4 "|"}' } #+end_src * Outdated ports The following ports are outdated: #+begin_src sh :session port-update :results output raw :exports results print_outdated_header port -q outdated | format_port_outdated #+end_src The following outdated ports are requested: #+begin_src sh :session port-update :results output raw :exports results print_outdated_header port -q outdated and requested | format_port_outdated #+end_src * Update log #+begin_src sh :session port-update :exports results :results verbatim for i in 1 2 3 4 5 6 7 8 9 0; do echo "."; done #+end_src
C-h v org-export-run-in-background
> Hi,
>
> is it possible to export an org file from the command line, so that a
> currently running Emacs instance is not disturbed? I want to export the
> attached org file and run the included source blocks, so I have an
> activity report in the end. I use the shell script pasted below, but
> there are two problems:
>
> - my Emacs instance is blocked during the execution of the shell
> scripts contained in the file
> - the script globally sets org-confirm-babel-evaluate to nil for my
> Emacs instance
>
> The second problem could possibly be solved with a local file variable.
> But the first problem remains. If I use emacs instead of emacsclient, it
> complains about a running Emacs instance.
>
> #!/bin/sh
>
> emacsclient -c \
> --eval "(progn
> (find-file \"macports.org\"))
> (setq org-confirm-babel-evaluate nil)
> (org-export-as-html 3)
> (kill-buffer)
> (delete-frame))"
>
> I'm using Org-Mode 7.7.
>
> Cheers,
> Viktor
>
--
Hi,
Sebastien Vauban wrote:
> #+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700
Where is the :engine directive documented?
Cheers,
Viktor
Viktor Rosenfeld <listuser36@googlemail.com> wrote:
> Hi,
>
> Sebastien Vauban wrote:
>
> > #+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700
>
> Where is the :engine directive documented?
>
The only place I know of is lisp/ob-sql.el - there is support for the following engines currently:
,----
| ...
| ('msosql (format "osql %s -s \"\t\" -i %s -o %s"
| (or cmdline "")
| (org-babel-process-file-name in-file)
| (org-babel-process-file-name out-file)))
| ('mysql (format "mysql %s < %s > %s"
| (or cmdline "")
| (org-babel-process-file-name in-file)
| (org-babel-process-file-name out-file)))
| ('postgresql (format
| "psql -A -P footer=off -F \"\t\" -f %s -o %s %s"
| (org-babel-process-file-name in-file)
| (org-babel-process-file-name out-file)
| (or cmdline "")))
| (t (error "no support for the %s sql engine" engine)))))
| ...
`----
Nick
Cool, thanks!
Nick Dokos wrote:
> Viktor Rosenfeld <listuser36@googlemail.com> wrote:
>
> > Hi,
> >
> > Sebastien Vauban wrote:
> >
> > > #+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700
> >
> > Where is the :engine directive documented?
> >
>
> The only place I know of is lisp/ob-sql.el - there is support for the following engines currently:
>
>
> ,----
> | ...
> | ('msosql (format "osql %s -s \"\t\" -i %s -o %s"
> | (or cmdline "")
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)))
> | ('mysql (format "mysql %s < %s > %s"
> | (or cmdline "")
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)))
> | ('postgresql (format
> | "psql -A -P footer=off -F \"\t\" -f %s -o %s %s"
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)
> | (or cmdline "")))
> | (t (error "no support for the %s sql engine" engine)))))
> | ...
> `----
>
> Nick
>
>
Hi Viktor and Nick,
Nick Dokos wrote:
> Viktor Rosenfeld <listuser36-gM/Ye1E23mwN+BqQ9rBEUg@public.gmane.org> wrote:
>> Sebastien Vauban wrote:
>>
>> > #+BABEL: :engine msosql :cmdline -S <SERVER> -U <USER> -P <PASS> -d <DATABASE> -n -w 700
>>
>> Where is the :engine directive documented?
>
> The only place I know of is lisp/ob-sql.el - there is support for the following engines currently:
>
>
> ,----
> | ...
> | ('msosql (format "osql %s -s \"\t\" -i %s -o %s"
> | (or cmdline "")
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)))
> | ('mysql (format "mysql %s < %s > %s"
> | (or cmdline "")
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)))
> | ('postgresql (format
> | "psql -A -P footer=off -F \"\t\" -f %s -o %s %s"
> | (org-babel-process-file-name in-file)
> | (org-babel-process-file-name out-file)
> | (or cmdline "")))
> | (t (error "no support for the %s sql engine" engine)))))
> | ...
> `----
Nothing to add. The `engine' directive is SQL-specific.
Note that I have in mind to add such an `engine' directive to Ledger as well:
being able to let the transactions be processed by `ledger', `hledger' or
`beancount' -- the ones I know of, and which are mostly compatible between
each other.
Best regards,
Seb
--
Sebastien Vauban
"Sebastien Vauban" <wxhgmqzgwmuf@spammotel.com> writes: > Hi Eric, > > Eric Schulte wrote: >>> Babel seems to interpret every *leading space* as *one empty column*. >>> Normal, feature, bug? >>> >>> Is there some workaround to this? I thought stating "scalar" would really >>> completely override any interpretation... >> >> I've just pushed up a fix which should resolve this issue. > > It does better things, but at least at the wrong place. > oh, I forgot to insert into a temporary buffer. This should now be fixed. Thanks -- Eric -- Eric Schulte http://cs.unm.edu/~eschulte/
Hi, Jambunathan K wrote: > > C-h v org-export-run-in-background This only works for org-export, but not for org-export-as-XXX. Additionally, it appears that setting org-confirm-babel-evaluate locally does not have an effect on background exports. Even setting it globally in a running Emacs instance won't work; the command has to be in a file that is loaded during Emacs startup. The help contains the following text which I don't understand: This variable is safe as a file local variable if its value satisfies the predicate which is byte-compiled expression. Cheers, Viktor > > > Hi, > > > > is it possible to export an org file from the command line, so that a > > currently running Emacs instance is not disturbed? I want to export the > > attached org file and run the included source blocks, so I have an > > activity report in the end. I use the shell script pasted below, but > > there are two problems: > > > > - my Emacs instance is blocked during the execution of the shell > > scripts contained in the file > > - the script globally sets org-confirm-babel-evaluate to nil for my > > Emacs instance > > > > The second problem could possibly be solved with a local file variable. > > But the first problem remains. If I use emacs instead of emacsclient, it > > complains about a running Emacs instance. > > > > #!/bin/sh > > > > emacsclient -c \ > > --eval "(progn > > (find-file \"macports.org\")) > > (setq org-confirm-babel-evaluate nil) > > (org-export-as-html 3) > > (kill-buffer) > > (delete-frame))" > > > > I'm using Org-Mode 7.7. > > > > Cheers, > > Viktor > > > > -- >
Viktor Rosenfeld <listuser36@googlemail.com> wrote:
> Hi,
>
> Jambunathan K wrote:
>
> >
> > C-h v org-export-run-in-background
>
> This only works for org-export, but not for org-export-as-XXX.
>
> Additionally, it appears that setting org-confirm-babel-evaluate locally
> does not have an effect on background exports. Even setting it globally
> in a running Emacs instance won't work; the command has to be in a file
> that is loaded during Emacs startup.
>
> The help contains the following text which I don't understand:
>
> This variable is safe as a file local variable if its value
> satisfies the predicate which is byte-compiled expression.
>
> Cheers,
> Viktor
>
> >
> > > Hi,
> > >
> > > is it possible to export an org file from the command line, so that a
> > > currently running Emacs instance is not disturbed? I want to export the
> > > attached org file and run the included source blocks, so I have an
> > > activity report in the end. I use the shell script pasted below, but
> > > there are two problems:
> > >
> > > - my Emacs instance is blocked during the execution of the shell
> > > scripts contained in the file
> > > - the script globally sets org-confirm-babel-evaluate to nil for my
> > > Emacs instance
> > >
> > > The second problem could possibly be solved with a local file variable.
> > > But the first problem remains. If I use emacs instead of emacsclient, it
> > > complains about a running Emacs instance.
> > >
> > > #!/bin/sh
> > >
> > > emacsclient -c \
> > > --eval "(progn
> > > (find-file \"macports.org\"))
> > > (setq org-confirm-babel-evaluate nil)
> > > (org-export-as-html 3)
> > > (kill-buffer)
> > > (delete-frame))"
> > >
> > > I'm using Org-Mode 7.7.
> > >
The usual method is to run a separate emacs in batch mode: that will avoid any conflicts
with the running instance. But batch implies -q, so you will have to provide a minimal .emacs
file that sets up enough structure to enable you to do what you want:
emacs --batch -l ~/minimal.emacs \
--eval '(let ((org-confirm-babel-evaluate nil))
(find-file "macports.org")
(org-export-as-html 3))'
should do it.
minimal.emacs would just load org - modulo path differences, something like:
--8<---------------cut here---------------start------------->8---
;;; -*- mode: emacs-lisp -*-
;;; constant part
(add-to-list 'load-path (expand-file-name "~/src/emacs/org/org-mode/lisp"))
(add-to-list 'load-path (expand-file-name "~/src/emacs/org/org-mode/contrib/lisp"))
(add-to-list 'auto-mode-alist '("\\.\\(org\\|org_archive\\|txt\\)$" . org-mode))
(require 'org-install)
--8<---------------cut here---------------end--------------->8---
Also note the let bind of org-confirm-babel-evaluate: that would alleviate the
second problem you mentioned above in the emacsclient case (and although it's
irrelevant in the emacs case, I still prefer the let bind over the explicit setq).
HTH,
Nick
While testing my response to Viktor's question, I ran into a problem. I used a test file that is slightly modified from a previous post of Tom Dye's: --8<---------------cut here---------------start------------->8--- * R tables #+TBLNAME: tbl-1 | column1 | column2 | |---------+---------| | 45 | 34 | | 77 | 56 | #+tblname: tbl-2 | col1 | col2 | |------+------| | a | b | | c | d | #+tblname: tbl-3 | c1 | c2 | |----+----| | A | B | | C | D | #+BEGIN_SRC R :var x=tbl-1 :var y=tbl-2 :var z=tbl-3 :colnames yes :exports both :results value z #+END_SRC --8<---------------cut here---------------end--------------->8--- Evaluating the code block correctly produces the result ,---- | | #+results: | | c1 | c2 | | |----+----| | | A | B | | | C | D | `---- but exporting (to ascii, PDF, HTML or ODT) chops off the first row of the result. For example, here is the ascii: ,---- | | c1 c2 | ----+---- | C D `---- HTML produces: ,---- | <pre class="example"> | c1 c2 | 1 C D | </pre> `---- Latex: ,---- | \begin{center} | \begin{tabular}{ll} | c1 & c2 \\ | \hline | C & D \\ | \end{tabular} | \end{center} `---- ODT: ,---- | <table:table table:name="" table:style-name="OrgTable"> | <table:table-column table:style-name="OrgTableColumn"/> | <table:table-column table:style-name="OrgTableColumn"/> | | <table:table-header-rows> | <table:table-row><table:table-cell table:style-name="OrgTblCellT"><text:p text:style-name="OrgTableHeadingLeft">c1</text:p></table:table-cell> | | <table:table-cell table:style-name="OrgTblCellT"><text:p text:style-name="OrgTableHeadingLeft">c2</text:p></table:table-cell> | </table:table-row> | | </table:table-header-rows> | | <table:table-rows> | <table:table-row><table:table-cell table:style-name="OrgTblCellTB"><text:p text:style-name="OrgTableContentsLeft">C</text:p></table:table-cell> | | <table:table-cell table:style-name="OrgTblCellTB"><text:p text:style-name="OrgTableContentsLeft">D</text:p></table:table-cell> | </table:table-row> | | </table:table-rows> | | </table:table> `---- Versions: GNU Emacs 24.0.50.2 (x86_64-unknown-linux-gnu, GTK+ Version 2.22.0) of 2011-09-13 Org-mode version 7.7 (release_7.7.396.gfaaa) Nick
Hi Nick,
Nick Dokos wrote:
> The usual method is to run a separate emacs in batch mode: that will avoid any conflicts
> with the running instance. But batch implies -q, so you will have to provide a minimal .emacs
> file that sets up enough structure to enable you to do what you want:
>
> emacs --batch -l ~/minimal.emacs \
> --eval '(let ((org-confirm-babel-evaluate nil))
> (find-file "macports.org")
> (org-export-as-html 3))'
>
> should do it.
Thanks for your help. I got it running using a similar setup. One more
question, it doesn't matter if I put the code in the file that is loaded
via -l or in the --eval block, correct? I've included the code in the my
org file, so I can tangle it. I find it better to have all the Emacs
code in one location.
Here's what I came up with:
(add-to-list 'load-path (expand-file-name "~/unix/src/org-mode/lisp"))
(add-to-list 'load-path (expand-file-name "~/unix/src/org-mode/contrib/lisp"))
(require 'org)
(require 'ob-sh)
(require 'ansi-color)
(let ((ansi-color-for-comint-mode nil)
(org-confirm-babel-evaluate nil)
(org-use-sub-superscripts nil))
(find-file "~/org/projects/macports/macports.org")
(org-with-point-at
(org-id-find "83583083-47B7-44DF-8474-1C6D03491C97" 'marker)
(org-babel-execute-subtree))
(org-export-as-html t)
(kill-buffer))
Thanks again,
Viktor
Viktor Rosenfeld <listuser36@googlemail.com> wrote: > ... > One more question, it doesn't matter if I put the code in the > file that is loaded via -l or in the --eval block, correct? I've > included the code in the my org file, so I can tangle it. I find it > better to have all the Emacs code in one location. > Correct. Which is preferable is very much a matter of usage patterns and taste. > Here's what I came up with: > > (add-to-list 'load-path (expand-file-name "~/unix/src/org-mode/lisp")) > (add-to-list 'load-path (expand-file-name "~/unix/src/org-mode/contrib/lisp")) > (require 'org) > (require 'ob-sh) > (require 'ansi-color) > (let ((ansi-color-for-comint-mode nil) > (org-confirm-babel-evaluate nil) > (org-use-sub-superscripts nil)) > (find-file "~/org/projects/macports/macports.org") > (org-with-point-at > (org-id-find "83583083-47B7-44DF-8474-1C6D03491C97" 'marker) > (org-babel-execute-subtree)) > (org-export-as-html t) > (kill-buffer)) You don't need the kill-buffer if emacs is just going to exit. But it doesn't hurt. Nick
Hi Eric,
Eric Schulte wrote:
> "Sebastien Vauban" <wxhgmqzgwmuf-geNee64TY+gS+FvcfC7Uqw@public.gmane.org> writes:
>> Eric Schulte wrote:
>>>> Babel seems to interpret every *leading space* as *one empty column*.
>>>> Normal, feature, bug?
>>>>
>>>> Is there some workaround to this? I thought stating "scalar" would really
>>>> completely override any interpretation...
>>>
>>> I've just pushed up a fix which should resolve this issue.
>>
>> It does better things, but at least at the wrong place.
>
> oh, I forgot to insert into a temporary buffer. This should now be
> fixed.
This works like a charm. Thanks a lot!
Best regards,
Seb
--
Sebastien Vauban
Nick Dokos <nicholas.dokos@hp.com> writes: > While testing my response to Viktor's question, I ran into a problem. > I used a test file that is slightly modified from a previous post of Tom Dye's: > > * R tables > > #+TBLNAME: tbl-1 > | column1 | column2 | > |---------+---------| > | 45 | 34 | > | 77 | 56 | > > #+tblname: tbl-2 > | col1 | col2 | > |------+------| > | a | b | > | c | d | > > #+tblname: tbl-3 > | c1 | c2 | > |----+----| > | A | B | > | C | D | > > #+BEGIN_SRC R :var x=tbl-1 :var y=tbl-2 :var z=tbl-3 :colnames yes :exports both :results value > z > #+END_SRC > > > > Evaluating the code block correctly produces the result > > ,---- > | > | #+results: > | | c1 | c2 | > | |----+----| > | | A | B | > | | C | D | > `---- > > but exporting (to ascii, PDF, HTML or ODT) chops off the first row of > the result. For example, here > is the ascii: > > ,---- > | > | c1 c2 > | ----+---- > | C D > `---- > > HTML produces: > > ,---- > | <pre class="example"> > | c1 c2 > | 1 C D > | </pre> > `---- > > Latex: > > ,---- > | \begin{center} > | \begin{tabular}{ll} > | c1 & c2 \\ > | \hline > | C & D \\ > | \end{tabular} > | \end{center} > `---- > > ODT: > > ,---- > | <table:table table:name="" table:style-name="OrgTable"> > | <table:table-column table:style-name="OrgTableColumn"/> > | <table:table-column table:style-name="OrgTableColumn"/> > | > | <table:table-header-rows> > | <table:table-row><table:table-cell > | table:style-name="OrgTblCellT"><text:p > | text:style-name="OrgTableHeadingLeft">c1</text:p></table:table-cell> > | > | <table:table-cell table:style-name="OrgTblCellT"><text:p > | text:style-name="OrgTableHeadingLeft">c2</text:p></table:table-cell> > | </table:table-row> > | > | </table:table-header-rows> > | > | <table:table-rows> > | <table:table-row><table:table-cell > | table:style-name="OrgTblCellTB"><text:p > | text:style-name="OrgTableContentsLeft">C</text:p></table:table-cell> > | > | <table:table-cell table:style-name="OrgTblCellTB"><text:p > | text:style-name="OrgTableContentsLeft">D</text:p></table:table-cell> > | </table:table-row> > | > | </table:table-rows> > | > | </table:table> > `---- > > Versions: > GNU Emacs 24.0.50.2 (x86_64-unknown-linux-gnu, GTK+ Version 2.22.0) of > 2011-09-13 > Org-mode version 7.7 (release_7.7.396.gfaaa) > > Nick > > Aloha Nick, I see the same behavior. Tom -- Thomas S. Dye http://www.tsdye.com
Thomas S. Dye <tsd@tsdye.com> wrote:
> Aloha Nick,
>
> I see the same behavior.
>
Thanks for confirming!
I'm not entirely sure but it seems to be R-specific: when babel does
variable assignments in org-babel-R-assign-elisp, it creates temp files
/tmp/babel-XXX/R-import-YYY and when the tables are written out to the
files, the first row of each table seems to go missing.
Nick
Nick Dokos <nicholas.dokos@hp.com> wrote:
> Thomas S. Dye <tsd@tsdye.com> wrote:
>
>
> > Aloha Nick,
> >
> > I see the same behavior.
> >
>
> Thanks for confirming!
>
> I'm not entirely sure but it seems to be R-specific: when babel does
> variable assignments in org-babel-R-assign-elisp, it creates temp files
> /tmp/babel-XXX/R-import-YYY and when the tables are written out to the
> files, the first row of each table seems to go missing.
>
I spoke too soon: I don't know what's going on, but the above is probably
wrong - please disregard. I'll take another whack at it over the weekend
(maybe).
Nick
Thomas S. Dye <tsd@tsdye.com> wrote:
> Nick Dokos <nicholas.dokos@hp.com> writes:
>
> > While testing my response to Viktor's question, I ran into a problem.
> > I used a test file that is slightly modified from a previous post of Tom Dye's:
> >
> > * R tables
> >
> > #+TBLNAME: tbl-1
> > | column1 | column2 |
> > |---------+---------|
> > | 45 | 34 |
> > | 77 | 56 |
> >
> > #+tblname: tbl-2
> > | col1 | col2 |
> > |------+------|
> > | a | b |
> > | c | d |
> >
> > #+tblname: tbl-3
> > | c1 | c2 |
> > |----+----|
> > | A | B |
> > | C | D |
> >
> > #+BEGIN_SRC R :var x=tbl-1 :var y=tbl-2 :var z=tbl-3 :colnames yes :exports both :results value
> > z
> > #+END_SRC
> >
> >
> >
> > Evaluating the code block correctly produces the result
> >
> > ,----
> > |
> > | #+results:
> > | | c1 | c2 |
> > | |----+----|
> > | | A | B |
> > | | C | D |
> > `----
> >
> > but exporting (to ascii, PDF, HTML or ODT) chops off the first row of
> > the result. For example, here
> > is the ascii:
> >
> > ,----
> > |
> > | c1 c2
> > | ----+----
> > | C D
> > `----
> > ...
>
> Aloha Nick,
>
> I see the same behavior.
>
Here's a progress report on this: the call chain is
,----
|
| org-babel-process-params(((:comments . "") (:shebang . "") (:cache . "no") (:padline . "") (:noweb . "no") (:tangle . "no") (:exports . "both") (:results . "replace value") (:var x (45 34) (77 56)) (:var y ("a" "b") ("c" "d")) (:var z ("A" "B") ("C" "D")) (:session . "none") (:rowname-names) (:result-type . value) (:result-params "replace" "value") (:padnewline . "yes") (:hlines . "no") (:colnames . "yes") (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" "c2"))))
| ...
| (org-babel-exp-in-export-file lang (org-babel-process-params (org-babel-merge-params (nth 2 info) (\` ((:results \, (if silent "silent" "replace")))))))
| ...
| org-babel-exp-results(("R" "z" ((:cache . "no") (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames . "yes") (:comments . "") (:exports . "both") (:hlines . "no") (:noweb . "no") (:padline . "") (:padnewline . "yes") (:result-params "replace" "value") (:result-type . value) (:results . "replace value") (:rowname-names) (:session . "none") (:shebang . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block nil "6443d36b0252c9e3192f813429018760bee3c11d")
| ...
| org-babel-exp-do-export(("R" "z" ((:cache . "no") (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames . "yes") (:comments . "") (:exports . "both") (:hlines . "no") (:noweb . "no") (:padline . "") (:padnewline . "yes") (:result-params "replace" "value") (:result-type . value) (:results . "replace value") (:rowname-names) (:session . "none") (:shebang . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block "6443d36b0252c9e3192f813429018760bee3c11d")
|
| org-babel-exp-src-block(...)
|
| org-export-blocks-preprocess()
| ...
| org-export-preprocess-string(...)
| ...
| org-export-as-ascii(nil nil nil "*Org ASCII Export*")
| org-export-as-ascii-to-buffer(nil)
| ...
| org-export(nil)
`----
So far, so good: x, y and z are still 2x2 tables. But evaluating the last form, we get
((:var x (77 56)) (:var y ("c" "d")) (:var z ("C" "D")) (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:rowname-names) (:result-params "replace" "value" "replace" "value") (:result-type . value) (:comments . "") (:shebang . "") (:cache . "no") (:padline . "") (:noweb . "no") ...)
The first row of each table has disappeared. Looks like a problem with org-babel-process-params.
Eric, can you take a look?
Thanks,
Nick
I turns out the problem here is that the table was being parsed for column headers twice, so the top two rows were both being taken as column headers. I've just pushed up a fix. Best -- Eric Nick Dokos <nicholas.dokos@hp.com> writes: > Thomas S. Dye <tsd@tsdye.com> wrote: > >> Nick Dokos <nicholas.dokos@hp.com> writes: >> >> > While testing my response to Viktor's question, I ran into a problem. >> > I used a test file that is slightly modified from a previous post of Tom Dye's: >> > >> > * R tables >> > >> > #+TBLNAME: tbl-1 >> > | column1 | column2 | >> > |---------+---------| >> > | 45 | 34 | >> > | 77 | 56 | >> > >> > #+tblname: tbl-2 >> > | col1 | col2 | >> > |------+------| >> > | a | b | >> > | c | d | >> > >> > #+tblname: tbl-3 >> > | c1 | c2 | >> > |----+----| >> > | A | B | >> > | C | D | >> > >> > #+BEGIN_SRC R :var x=tbl-1 :var y=tbl-2 :var z=tbl-3 :colnames yes :exports both :results value >> > z >> > #+END_SRC >> > >> > >> > >> > Evaluating the code block correctly produces the result >> > >> > ,---- >> > | >> > | #+results: >> > | | c1 | c2 | >> > | |----+----| >> > | | A | B | >> > | | C | D | >> > `---- >> > >> > but exporting (to ascii, PDF, HTML or ODT) chops off the first row of >> > the result. For example, here >> > is the ascii: >> > >> > ,---- >> > | >> > | c1 c2 >> > | ----+---- >> > | C D >> > `---- >> > ... >> >> Aloha Nick, >> >> I see the same behavior. >> > > Here's a progress report on this: the call chain is > > ,---- > | > | org-babel-process-params(((:comments . "") (:shebang . "") (:cache > | . "no") (:padline . "") (:noweb . "no") (:tangle . "no") (:exports > | . "both") (:results . "replace value") (:var x (45 34) (77 56)) > | (:var y ("a" "b") ("c" "d")) (:var z ("A" "B") ("C" "D")) (:session > | . "none") (:rowname-names) (:result-type . value) (:result-params > | "replace" "value") (:padnewline . "yes") (:hlines . "no") (:colnames > | . "yes") (:colname-names (x "column1" "column2") (y "col1" "col2") > | (z "c1" "c2")))) > | ... > | (org-babel-exp-in-export-file lang (org-babel-process-params (org-babel-merge-params (nth 2 info) (\` ((:results \, (if silent "silent" "replace"))))))) > | ... > | org-babel-exp-results(("R" "z" ((:cache . "no") (:colname-names (x > | "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames > | . "yes") (:comments . "") (:exports . "both") (:hlines . "no") > | (:noweb . "no") (:padline . "") (:padnewline . "yes") > | (:result-params "replace" "value") (:result-type . value) (:results > | . "replace value") (:rowname-names) (:session . "none") (:shebang > | . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") > | ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block nil > | "6443d36b0252c9e3192f813429018760bee3c11d") > | ... > | org-babel-exp-do-export(("R" "z" ((:cache . "no") (:colname-names > | (x "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames > | . "yes") (:comments . "") (:exports . "both") (:hlines . "no") > | (:noweb . "no") (:padline . "") (:padnewline . "yes") > | (:result-params "replace" "value") (:result-type . value) (:results > | . "replace value") (:rowname-names) (:session . "none") (:shebang > | . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") > | ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block > | "6443d36b0252c9e3192f813429018760bee3c11d") > | > | org-babel-exp-src-block(...) > | > | org-export-blocks-preprocess() > | ... > | org-export-preprocess-string(...) > | ... > | org-export-as-ascii(nil nil nil "*Org ASCII Export*") > | org-export-as-ascii-to-buffer(nil) > | ... > | org-export(nil) > `---- > > So far, so good: x, y and z are still 2x2 tables. But evaluating the last form, we get > > ((:var x (77 56)) (:var y ("c" "d")) (:var z ("C" "D")) > (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" > "c2")) (:rowname-names) (:result-params "replace" "value" "replace" > "value") (:result-type . value) (:comments . "") (:shebang . "") > (:cache . "no") (:padline . "") (:noweb . "no") ...) > > The first row of each table has disappeared. Looks like a problem with org-babel-process-params. > Eric, can you take a look? > > Thanks, > Nick > -- Eric Schulte http://cs.unm.edu/~eschulte/
Eric Schulte <schulte.eric@gmail.com> wrote: > I turns out the problem here is that the table was being parsed for > column headers twice, so the top two rows were both being taken as > column headers. I've just pushed up a fix. > Works fine - thanks! > Best -- Eric > > Nick Dokos <nicholas.dokos@hp.com> writes: > > > Thomas S. Dye <tsd@tsdye.com> wrote: > > > >> Nick Dokos <nicholas.dokos@hp.com> writes: > >> > >> > While testing my response to Viktor's question, I ran into a problem. > >> > I used a test file that is slightly modified from a previous post of Tom Dye's: > >> > > >> > * R tables > >> > > >> > #+TBLNAME: tbl-1 > >> > | column1 | column2 | > >> > |---------+---------| > >> > | 45 | 34 | > >> > | 77 | 56 | > >> > > >> > #+tblname: tbl-2 > >> > | col1 | col2 | > >> > |------+------| > >> > | a | b | > >> > | c | d | > >> > > >> > #+tblname: tbl-3 > >> > | c1 | c2 | > >> > |----+----| > >> > | A | B | > >> > | C | D | > >> > > >> > #+BEGIN_SRC R :var x=tbl-1 :var y=tbl-2 :var z=tbl-3 :colnames yes :exports both :results value > >> > z > >> > #+END_SRC > >> > > >> > > >> > > >> > Evaluating the code block correctly produces the result > >> > > >> > ,---- > >> > | > >> > | #+results: > >> > | | c1 | c2 | > >> > | |----+----| > >> > | | A | B | > >> > | | C | D | > >> > `---- > >> > > >> > but exporting (to ascii, PDF, HTML or ODT) chops off the first row of > >> > the result. For example, here > >> > is the ascii: > >> > > >> > ,---- > >> > | > >> > | c1 c2 > >> > | ----+---- > >> > | C D > >> > `---- > >> > ... > >> > >> Aloha Nick, > >> > >> I see the same behavior. > >> > > > > Here's a progress report on this: the call chain is > > > > ,---- > > | > > | org-babel-process-params(((:comments . "") (:shebang . "") (:cache > > | . "no") (:padline . "") (:noweb . "no") (:tangle . "no") (:exports > > | . "both") (:results . "replace value") (:var x (45 34) (77 56)) > > | (:var y ("a" "b") ("c" "d")) (:var z ("A" "B") ("C" "D")) (:session > > | . "none") (:rowname-names) (:result-type . value) (:result-params > > | "replace" "value") (:padnewline . "yes") (:hlines . "no") (:colnames > > | . "yes") (:colname-names (x "column1" "column2") (y "col1" "col2") > > | (z "c1" "c2")))) > > | ... > > | (org-babel-exp-in-export-file lang (org-babel-process-params (org-babel-merge-params (nth 2 info) (\` ((:results \, (if silent "silent" "replace"))))))) > > | ... > > | org-babel-exp-results(("R" "z" ((:cache . "no") (:colname-names (x > > | "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames > > | . "yes") (:comments . "") (:exports . "both") (:hlines . "no") > > | (:noweb . "no") (:padline . "") (:padnewline . "yes") > > | (:result-params "replace" "value") (:result-type . value) (:results > > | . "replace value") (:rowname-names) (:session . "none") (:shebang > > | . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") > > | ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block nil > > | "6443d36b0252c9e3192f813429018760bee3c11d") > > | ... > > | org-babel-exp-do-export(("R" "z" ((:cache . "no") (:colname-names > > | (x "column1" "column2") (y "col1" "col2") (z "c1" "c2")) (:colnames > > | . "yes") (:comments . "") (:exports . "both") (:hlines . "no") > > | (:noweb . "no") (:padline . "") (:padnewline . "yes") > > | (:result-params "replace" "value") (:result-type . value) (:results > > | . "replace value") (:rowname-names) (:session . "none") (:shebang > > | . "") (:tangle . "no") (:var x (45 34) (77 56)) (:var y ("a" "b") > > | ("c" "d")) (:var z ("A" "B") ("C" "D"))) "" nil 0) block > > | "6443d36b0252c9e3192f813429018760bee3c11d") > > | > > | org-babel-exp-src-block(...) > > | > > | org-export-blocks-preprocess() > > | ... > > | org-export-preprocess-string(...) > > | ... > > | org-export-as-ascii(nil nil nil "*Org ASCII Export*") > > | org-export-as-ascii-to-buffer(nil) > > | ... > > | org-export(nil) > > `---- > > > > So far, so good: x, y and z are still 2x2 tables. But evaluating the last form, we get > > > > ((:var x (77 56)) (:var y ("c" "d")) (:var z ("C" "D")) > > (:colname-names (x "column1" "column2") (y "col1" "col2") (z "c1" > > "c2")) (:rowname-names) (:result-params "replace" "value" "replace" > > "value") (:result-type . value) (:comments . "") (:shebang . "") > > (:cache . "no") (:padline . "") (:noweb . "no") ...) > > > > The first row of each table has disappeared. Looks like a problem with org-babel-process-params. > > Eric, can you take a look? > > > > Thanks, > > Nick > > > > -- > Eric Schulte > http://cs.unm.edu/~eschulte/ >
Hi, I'm trying to copy the values of a column in one table to another table in. If I understand the manual correctly, this should be possible with remote references. But I can't get it to work. In the example below, if I evaluate the table formula, column 2 of table "bar" is copied into column 3. Cheers, Viktor * Source table #+TBLNAME: foo | Query | Time | |-------+------| | 1 | 123 | | 2 | 234 | | 3 | 345 | * Reference table #+TBLNAME: bar | Query | Garbage | Old Time | |-------+---------+----------| | 1 | a | | | 2 | b | | | 3 | c | | #+TBLFM: $3=remote(foo,$2) * Expected | Query | Garbage | Old Time | |-------+---------+----------| | 1 | a | 123 | | 2 | b | 234 | | 3 | c | 345 | * Produced | Query | Garbage | Old Time | |-------+---------+----------| | 1 | a | a | | 2 | b | b | | 3 | c | c |
Viktor Rosenfeld <listuser36@googlemail.com> wrote: > Hi, > > I'm trying to copy the values of a column in one table to another table > in. If I understand the manual correctly, this should be possible with > remote references. But I can't get it to work. In the example below, if > I evaluate the table formula, column 2 of table "bar" is copied into > column 3. > > Cheers, > Viktor > > * Source table > > #+TBLNAME: foo > | Query | Time | > |-------+------| > | 1 | 123 | > | 2 | 234 | > | 3 | 345 | > > * Reference table > > #+TBLNAME: bar > | Query | Garbage | Old Time | > |-------+---------+----------| > | 1 | a | | > | 2 | b | | > | 3 | c | | > #+TBLFM: $3=remote(foo,$2) > The manual says: $3 = remote(FOO, @@#$2) copy column 2 from table FOO into column 3 of the current table and that works. Nick > * Expected > > | Query | Garbage | Old Time | > |-------+---------+----------| > | 1 | a | 123 | > | 2 | b | 234 | > | 3 | c | 345 | > > * Produced > > | Query | Garbage | Old Time | > |-------+---------+----------| > | 1 | a | a | > | 2 | b | b | > | 3 | c | c | >
Nick Dokos <nicholas.dokos@hp.com> wrote:
> Viktor Rosenfeld <listuser36@googlemail.com> wrote:
>
> > Hi,
> >
> > I'm trying to copy the values of a column in one table to another table
> > in. If I understand the manual correctly, this should be possible with
> > remote references. But I can't get it to work. In the example below, if
> > I evaluate the table formula, column 2 of table "bar" is copied into
> > column 3.
> >
> > Cheers,
> > Viktor
> >
> > * Source table
> >
> > #+TBLNAME: foo
> > | Query | Time |
> > |-------+------|
> > | 1 | 123 |
> > | 2 | 234 |
> > | 3 | 345 |
> >
> > * Reference table
> >
> > #+TBLNAME: bar
> > | Query | Garbage | Old Time |
> > |-------+---------+----------|
> > | 1 | a | |
> > | 2 | b | |
> > | 3 | c | |
> > #+TBLFM: $3=remote(foo,$2)
> >
>
> The manual says:
>
> $3 = remote(FOO, @@#$2) copy column 2 from table FOO into
> column 3 of the current table
>
> and that works.
>
A quick look shows (modulo mistakes) that the difference is that the
second form matches org-table-range-regexp, whereas the first one does
not: I'm not sure if that changed recently and/or whether it's
deliberate.
Nick
Nick Dokos wrote:
> The manual says:
>
> $3 = remote(FOO, @@#$2) copy column 2 from table FOO into
> column 3 of the current table
>
> and that works.
Thank you, Nick. Wer lesen kann, ist klar im Vorteil. (Those who can read
are at a clear advantage -- a German saying that is appropriate here I
guess.)
Cheers,
Viktor
Dear all, I made a small refresh of the website. http://orgmode.org - random quote (taken from worg/org-quotes.org) - random screenshot - twitter feed - g+/flattr/fb buttons - french translation (http://orgmode.org/fr/) If you want to help with the translation in your language, please send your public key and I will give you push access. This is just a bunch of .org files, only 1 hour needed to translate the whole website. Hope you like it! -- Bastien
Bastien <bzg@altern.org> writes: > http://orgmode.org I think ever since the server moved the favicon has been missing. Could it be added back, please? Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ SD adaptation for Waldorf rackAttack V1.04R1: http://Synth.Stromeko.net/Downloads.html#WaldorfSDada
Looks beautiful. Unfortunately, it does not work with large fonts. Try setting Firefox to the maximum value for the minimum font size to see what I mean. You might or might not need to make the window size smaller also. Look at the bottom line on the screen, then scroll down once. Now look at the top line on the screen. You should notice that several lines are missing. The top bar obscures them. In other words, page down does not work correctly, because the top bar was designed for normal fonts only. Thanks. Samuel -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com === Bigotry against people with serious diseases is still bigotry.
Samuel Wales <samologist@gmail.com> writes: > Looks beautiful. +1 Thanks! The only thing I don't particularly care for is the twitter box. If I have the browser visible visiting this page, as I do right now, and especially on a separate monitor, I keep getting distracted by motion in my peripheral vision! It's almost worse because of the faint display of the box when not hovering over it. Minor point, mind you; I just don't like animations on web sites... others will of course disagree! > > Unfortunately, it does not work with large fonts. > > Try setting Firefox to the maximum value for the minimum font size to > see what I mean. You might or might not need to make the window size > smaller also. > > Look at the bottom line on the screen, then scroll down once. Now > look at the top line on the screen. > > You should notice that several lines are missing. The top bar obscures them. > > In other words, page down does not work correctly, because the top bar > was designed for normal fonts only. > > Thanks. > > Samuel Samuel, I'm not sure what you mean here. I'm using iceweasel 8.0 so am not sure what version of Firefox this is but if I increase the font size (via C-=) to the maximum size available, things still work mostly fine. The text in the title bar (Org: Your life...) does scroll over the quote bar but there are no problems scrolling back and forth otherwise in the main text section. -- : Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 24.0.92.1 : using Org-mode version 7.7 (release_7.7.381.g05ea.dirty)
Hi Eric, On 2011-12-13, Eric S Fraga <e.fraga@ucl.ac.uk> wrote: > the box when not hovering over it. Minor point, mind you; I just don't > like animations on web sites... others will of course disagree! Agreed, though I have animations turned off. Some find animations to be an accessibility issue. > Samuel, I'm not sure what you mean here. I'm using iceweasel 8.0 so am > not sure what version of Firefox this is but if I increase the font size > (via C-=) to the maximum size available, things still work mostly Go to preferences/content/advanced and set the maximum minimum font size. You might or might not need a smaller window or to set lower resolution. Probably don't need that unless the whole page appears, of course. Keep trying. You will reproduce. Samuel -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com === Bigotry against people with serious diseases is still bigotry.
Achim Gratz <Stromeko@nexgo.de> writes:
> Bastien <bzg@altern.org> writes:
>> http://orgmode.org
>
> I think ever since the server moved the favicon has been missing. Could
> it be added back, please?
Done - please confirm.
Thanks,
--
Bastien
Bastien <bzg@altern.org> wrote:
> Achim Gratz <Stromeko@nexgo.de> writes:
>
> > Bastien <bzg@altern.org> writes:
> >> http://orgmode.org
> >
> > I think ever since the server moved the favicon has been missing. Could
> > it be added back, please?
>
> Done - please confirm.
>
Never noticed its absence, but I can see it now :)
Hi Samuel, yes, there are definitely some readability and accessibility issues that need to be fixed. The css is here: http://orgmode.org/org.css If anyone wants to send improvement, please do! We can also use a different .css, the html file are quite simple. I hope you understand I'd rather not dedicate too much time on this (I did already), as I have so many remaining Org bugs to fix... but help is welcome ! Thanks for all your feedback, -- Bastien
Hi Eric, Eric S Fraga <e.fraga@ucl.ac.uk> writes: > The only thing I don't particularly care for is the twitter box. If I > have the browser visible visiting this page, as I do right now, and > especially on a separate monitor, I keep getting distracted by motion in > my peripheral vision! It's almost worse because of the faint display of > the box when not hovering over it. Minor point, mind you; I just don't > like animations on web sites... others will of course disagree! If people want to get rid of this box in the index* pages, no objection. But.. wait.. why do you keep the index.html page open then? I don't make oops-here-is-another-stupid-bugfix-releases *that* often :) > Samuel, I'm not sure what you mean here. I'm using iceweasel 8.0 so am > not sure what version of Firefox this is but if I increase the font size > (via C-=) to the maximum size available, things still work mostly > fine. I guess it depends on the screen resolution. I have a fairly small screen (on a fairly old computer) and I see what Samuel means... Best, -- Bastien
Hi Bastien, Thanks for reproducing. And thanks for the release. Then maybe the bar is specified in pixels instead of ems? Just a guess. Samuel -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com === Bigotry against people with serious diseases is still bigotry.
Bastien <bzg@altern.org> writes: > Done - please confirm. Requesting /favicon.ico succeeds, but gets me a file that is exactly 0 bytes. Asking for /org-mode-unicorn.ico gets a file that contains the following text: HTTP/1.1 200 OK Date: Sun, 02 Jan 2011 14:48:25 GMT Server: Apache Last-Modified: Thu, 18 Jun 2009 21:43:15 GMT ETag: "90e230f-47e-46ca64b90e2c0" Accept-Ranges: bytes Content-Length: 1150 Content-Type: image/x-icon The wayback archive has a correct version of that file: http://web.archive.org/web/20100925104250/http://orgmode.org/org-mode-unicorn.ico (and it should probably be linked to /favicon.ico for the benefit of older browsers). Firefox seems to prefer the alternate PNG version of the icon and so it does display. Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2: http://Synth.Stromeko.net/Downloads.html#WaldorfSDada
Hi,
I honestly liked the old one better. Some of my concrete criticism are:
- The old screenshot conveyed what org-mode was all about. Outlining
and project planning. The new screenshots take a lot longer to load
and even if you click on them, their content is not always accessible.
- I don't like this particular type writer font. On a Mac it's very thin
compared to the serifless font that makes up the normal text. But
that's just aesthetics.
- I don't like the fixed top bar with the Org-Mode motto and the random
quote. It takes up about 20% of vertical screen space of my laptop. A
complete waste in my view. The black background also does not cover
the text behind it -- one or two pixel of the text are visible above
the bar.
- The old homepage had all the easily accessible information on one
page. So you could go to the homepage, search for "manual" or "mailing
list" and go on from there. (I sometimes work like that.) The new
homepage only contains screenshots (that don't convey a lot of
meaningful information) and the org-mode mission statement. Useful
information is available but the links are somewhat hidden in the
description of what org is about and some are hard to find. (E.g. the
manual is linked with the text "documented extensively".) I believe
that people quickly scan web pages for relevant information and the
current format makes that hard.
Cheers,
Viktor
Bastien wrote:
> Dear all,
>
> I made a small refresh of the website.
>
> http://orgmode.org
>
> - random quote (taken from worg/org-quotes.org)
> - random screenshot
> - twitter feed
> - g+/flattr/fb buttons
> - french translation (http://orgmode.org/fr/)
>
> If you want to help with the translation in your language,
> please send your public key and I will give you push access.
> This is just a bunch of .org files, only 1 hour needed to
> translate the whole website.
>
> Hope you like it!
>
> --
> Bastien
>
Viktor Rosenfeld <listuser36@googlemail.com> wrote:
> - The old homepage had all the easily accessible information on one
> page. So you could go to the homepage, search for "manual" or "mailing
> list" and go on from there. (I sometimes work like that.) The new
> homepage only contains screenshots (that don't convey a lot of
> meaningful information) and the org-mode mission statement. Useful
> information is available but the links are somewhat hidden in the
> description of what org is about and some are hard to find. (E.g. the
> manual is linked with the text "documented extensively".)
At least on Firefox, there *is* a panel of links
(Home/Download/Documentation/Community/News/More) and the manual is
available under the "Documentation" link.
Nick
Hi Viktor, Viktor Rosenfeld <listuser36@googlemail.com> writes: > - The old screenshot conveyed what org-mode was all about. Outlining > and project planning. The new screenshots take a lot longer to load > and even if you click on them, their content is not always > accessible. I agree. Feel free to contribute to Worg with relevant screenshots that we can select as "featured on the index page." Also add title and caption for these screenshots. > - I don't like this particular type writer font. On a Mac it's very thin > compared to the serifless font that makes up the normal text. But > that's just aesthetics. > > - I don't like the fixed top bar with the Org-Mode motto and the random > quote. It takes up about 20% of vertical screen space of my laptop. A > complete waste in my view. The black background also does not cover > the text behind it -- one or two pixel of the text are visible above > the bar. Can you make an alternate .css with non-fixed top bar and a better font? > - The old homepage had all the easily accessible information on one > page. So you could go to the homepage, search for "manual" or "mailing > list" and go on from there. (I sometimes work like that.) The new > homepage only contains screenshots (that don't convey a lot of > meaningful information) and the org-mode mission statement. Useful > information is available but the links are somewhat hidden in the > description of what org is about and some are hard to find. (E.g. the > manual is linked with the text "documented extensively".) I believe > that people quickly scan web pages for relevant information and the > current format makes that hard. I somehow agree that it was good to have everything in one page, but I don't think that page was *that* readable. In any case, you can always pull the website (it's just a git repository) and you have everything in "one directory". If you have any patch that would make the website better to your eyes, please send it! Thanks, -- Bastien
Viktor Rosenfeld <listuser36@googlemail.com> writes: > Hi, > > I honestly liked the old one better. Some of my concrete criticism are: > > - The old screenshot conveyed what org-mode was all about. Outlining > and project planning. The new screenshots take a lot longer to load > and even if you click on them, their content is not always accessible. > +1 for preferring the old screenshot, for exactly those reasons mentioned above. While it is fun for me to look at random screenshots from Worg, part of the purpose of this site is as a gateway to Org-mode, and I fear many of these random screenshots are meaningless (or even intimidating) unless you already have a good understanding of Org-mode. > > - I don't like the fixed top bar with the Org-Mode motto and the random > quote. It takes up about 20% of vertical screen space of my laptop. A > complete waste in my view. The black background also does not cover > the text behind it -- one or two pixel of the text are visible above > the bar. > I also find that (in part because of the new larger screenshots) I can't see any content upon first loading the page (see this screenshot [1]). > > - The old homepage had all the easily accessible information on one > page. So you could go to the homepage, search for "manual" or "mailing > list" and go on from there. (I sometimes work like that.) The new > homepage only contains screenshots (that don't convey a lot of > meaningful information) and the org-mode mission statement. Useful > information is available but the links are somewhat hidden in the > description of what org is about and some are hard to find. (E.g. the > manual is linked with the text "documented extensively".) I agree that the manual should be linked with the term "manual", as (unfortunately IMO) it seems many Emacs and Org-mode users aren't comfortable using the build in Emacs info system and prefer to use the online manual exclusively. Aside from those items mentioned above I do generally like the style of the new site. Best, > I believe that people quickly scan web pages for relevant information > and the current format makes that hard. > > Cheers, > Viktor > > Bastien wrote: > >> Dear all, >> >> I made a small refresh of the website. >> >> http://orgmode.org >> >> - random quote (taken from worg/org-quotes.org) >> - random screenshot >> - twitter feed >> - g+/flattr/fb buttons >> - french translation (http://orgmode.org/fr/) >> >> If you want to help with the translation in your language, >> please send your public key and I will give you push access. >> This is just a bunch of .org files, only 1 hour needed to >> translate the whole website. >> >> Hope you like it! >> >> -- >> Bastien >> > Footnotes: [1] http://imgur.com/mL3zb -- Eric Schulte http://cs.unm.edu/~eschulte/
On 12/12/2011 06:39 PM, Bastien wrote:
> Dear all,
>
> I made a small refresh of the website.
>
> http://orgmode.org
>
> - random quote (taken from worg/org-quotes.org)
> - random screenshot
> - twitter feed
> - g+/flattr/fb buttons
> - french translation (http://orgmode.org/fr/)
>
> If you want to help with the translation in your language,
> please send your public key and I will give you push access.
> This is just a bunch of .org files, only 1 hour needed to
> translate the whole website.
>
> Hope you like it!
>
The random screenshot I saw was one of mine. :)
Scott Randby
Bastien <bzg@altern.org> writes: > Hi Eric, Hi Bastien! > > Eric S Fraga <e.fraga@ucl.ac.uk> writes: > >> The only thing I don't particularly care for is the twitter box. If I >> have the browser visible visiting this page, as I do right now, and >> especially on a separate monitor, I keep getting distracted by motion in >> my peripheral vision! It's almost worse because of the faint display of >> the box when not hovering over it. Minor point, mind you; I just don't >> like animations on web sites... others will of course disagree! > > If people want to get rid of this box in the index* pages, no objection. > > But.. wait.. why do you keep the index.html page open then? I don't > make oops-here-is-another-stupid-bugfix-releases *that* often :) Very true! I guess I was in a pedantic or picky mood last night... :( Ignore my rant. The site looks very nice otherwise! -- : Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 24.0.90.1 : using Org-mode version 7.8.02 (release_7.8.02.2.g490d6a)
Bastien <bzg@altern.org> writes: > Viktor Rosenfeld <listuser36@googlemail.com> writes: >> - The old homepage had all the easily accessible information on one >> page. So you could go to the homepage, search for "manual" or "mailing >> list" and go on from there. (I sometimes work like that.) The new >> homepage only contains screenshots (that don't convey a lot of >> meaningful information) and the org-mode mission statement. Useful >> information is available but the links are somewhat hidden in the >> description of what org is about and some are hard to find. (E.g. the >> manual is linked with the text "documented extensively".) I believe >> that people quickly scan web pages for relevant information and the >> current format makes that hard. > > I somehow agree that it was good to have everything in one page, but I > don't think that page was *that* readable. In any case, you can always I agree with Bastien here. In fact, I used to use the old website in the manner the new one requires: using the left index to go to the bit I wanted. The new one is more obvious in this regard. -- : Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D) in Emacs 24.0.90.1 : using Org-mode version 7.8.02 (release_7.8.02.2.g490d6a)
Hi Eric, Eric Schulte <eric.schulte@gmx.com> writes: > +1 for preferring the old screenshot, for exactly those reasons > mentioned above. While it is fun for me to look at random > screenshots from Worg, part of the purpose of this site is as a > gateway to Org-mode, and I fear many of these random screenshots are > meaningless (or even intimidating) unless you already have a good > understanding of Org-mode. Again, I agree. Patch welcome! >> - I don't like the fixed top bar with the Org-Mode motto and the random >> quote. It takes up about 20% of vertical screen space of my laptop. A >> complete waste in my view. The black background also does not cover >> the text behind it -- one or two pixel of the text are visible above >> the bar. > > I also find that (in part because of the new larger screenshots) I can't > see any content upon first loading the page (see this screenshot [1]). Mh.. your screen seems quite small. If you can fix the .css to display the website better on your screen, please do, I don't have time at hand now to do it myself. >> - The old homepage had all the easily accessible information on one >> page. So you could go to the homepage, search for "manual" or "mailing >> list" and go on from there. (I sometimes work like that.) The new >> homepage only contains screenshots (that don't convey a lot of >> meaningful information) and the org-mode mission statement. Useful >> information is available but the links are somewhat hidden in the >> description of what org is about and some are hard to find. (E.g. the >> manual is linked with the text "documented extensively".) > > I agree that the manual should be linked with the term "manual", as > (unfortunately IMO) it seems many Emacs and Org-mode users aren't > comfortable using the build in Emacs info system and prefer to use the > online manual exclusively. Well, the manual is two clicks away: 1) "Documentation" 2) "The online manual" I find it good to gather every available documentation in one single page, that people will later be exposed to when looking for "orgmode documentation". That's also the benefit of having one page. And one Google Search away: http://www.google.fr/search?&ie=UTF-8&q=orgmode+manual -> http://orgmode.org/org.html > Aside from those items mentioned above I do generally like the style of > the new site. Thanks! -- Bastien
Hi Bastien, Bastien wrote: > > - The old screenshot conveyed what org-mode was all about. Outlining > > and project planning. The new screenshots take a lot longer to load > > and even if you click on them, their content is not always > > accessible. > > I agree. Feel free to contribute to Worg with relevant screenshots > that we can select as "featured on the index page." Also add title > and caption for these screenshots. I think the old screenshot was good enough. A link beneath it to the screenshot page on Worg would be useful. > > - I don't like this particular type writer font. On a Mac it's very thin > > compared to the serifless font that makes up the normal text. But > > that's just aesthetics. > > > > - I don't like the fixed top bar with the Org-Mode motto and the random > > quote. It takes up about 20% of vertical screen space of my laptop. A > > complete waste in my view. The black background also does not cover > > the text behind it -- one or two pixel of the text are visible above > > the bar. > > Can you make an alternate .css with non-fixed top bar and a better font? Maybe on the weekend, but I can't promise anything. > > - The old homepage had all the easily accessible information on one > > page. So you could go to the homepage, search for "manual" or "mailing > > list" and go on from there. (I sometimes work like that.) The new > > homepage only contains screenshots (that don't convey a lot of > > meaningful information) and the org-mode mission statement. Useful > > information is available but the links are somewhat hidden in the > > description of what org is about and some are hard to find. (E.g. the > > manual is linked with the text "documented extensively".) I believe > > that people quickly scan web pages for relevant information and the > > current format makes that hard. > > I somehow agree that it was good to have everything in one page, but I > don't think that page was *that* readable. In any case, you can always > pull the website (it's just a git repository) and you have everything > in "one directory". I agree that the old page wasn't very readable as it did contain a lot of information. I just don't feel that the new layout is better, on the contrary. Cheers, Viktor
[-- Attachment #1: Type: text/plain, Size: 1676 bytes --] Dear Viktor, dear Bastien, On 13.12.2011, at 22:50, Viktor Rosenfeld wrote: > [...]- I don't like this particular type writer font. On a Mac it's very thin > compared to the serifless font that makes up the normal text. But > that's just aesthetics. > > - I don't like the fixed top bar with the Org-Mode motto and the random > quote. It takes up about 20% of vertical screen space of my laptop. A > complete waste in my view. The black background also does not cover > the text behind it -- one or two pixel of the text are visible above > the bar. I really like the new web site "look" (thanks, Bastien!): I tested it on a Mac with recent versions of Firefox, Safari and Opera and was (so far) very pleased with both usability and aesthetics. However, that was on a machine with a large screen and I can imagine that some layout and font decisions could be improved for smaller systems - it does not work so well on the iPhone where I also did some testing. Almost every web site has that problem - the solution usually is to provide more than one stylesheet and provide a suitable one depending on the browser identification; this works well with most SmartPhones and the iPad. We have some institute-internal services that do this (not difficult to implement with PHP) - is this an option for the current orgmode-site? Warm regards, Stefan -- Dr. Stefan Vollmar, Dipl.-Phys. Head of IT group Max-Planck-Institut für neurologische Forschung Gleuelerstr. 50, 50931 Köln, Germany Tel.: +49-221-4726-213 FAX +49-221-4726-298 Tel.: +49-221-478-5713 Mobile: 0160-93874279 Email: vollmar@nf.mpg.de http://www.nf.mpg.de [-- Attachment #2: smime.p7s --] [-- Type: application/pkcs7-signature, Size: 4409 bytes --]
Hi all, Old School Org screenshots are back in orgmode.org. Enjoy! -- Bastien
Viktor Rosenfeld <listuser36@googlemail.com> writes:
>> Can you make an alternate .css with non-fixed top bar and a better font?
>
> Maybe on the weekend, but I can't promise anything.
Thanks in advance for this!
--
Bastien
> >> +1 for preferring the old screenshot, for exactly those reasons >> mentioned above. While it is fun for me to look at random >> screenshots from Worg, part of the purpose of this site is as a >> gateway to Org-mode, and I fear many of these random screenshots are >> meaningless (or even intimidating) unless you already have a good >> understanding of Org-mode. > > Again, I agree. Patch welcome! > Ah, I should have known to hold my tongue! :) I pushed a patch up to a new "old-image" branch in the orgweb repository, but it seems you've now taken care of this independently so please feel free to delete that branch. > >>> - I don't like the fixed top bar with the Org-Mode motto and the random >>> quote. It takes up about 20% of vertical screen space of my laptop. A >>> complete waste in my view. The black background also does not cover >>> the text behind it -- one or two pixel of the text are visible above >>> the bar. >> >> I also find that (in part because of the new larger screenshots) I can't >> see any content upon first loading the page (see this screenshot [1]). > > Mh.. your screen seems quite small. If you can fix the .css to display > the website better on your screen, please do, I don't have time at hand > now to do it myself. > One of my preferred layout looks something like the following, which can lead to a small windows for Firefox. +----------------+---------------+ | | | | | | | Firefox | Emacs | | | | | | | +----------------+---------------+ | Terminal | +--------------------------------+ There is a huge difference in readability between those websites which take variable window size into consideration and those which assume a large screen (or a full screen browser). I'll take a shot at changing the CSS in the orgweb repository to facility smaller layouts (I guess some special CSS for mobile browsers may also make sense). ...looking... It looks like making the Org-mode website responsive to the viewers window size should be as simple as adding a couple of "@media" guards [1] to the css page. I may have time to tackle this over the weekend. I do need some help compiling the Org-mode web page from the git repository. Is there an org-mode publishing project which I should define locally? In general how are the many file in orgweb compiled into the site? Thanks, -- Eric Schulte http://cs.unm.edu/~eschulte/ Footnotes: [1] http://www.w3.org/TR/css3-mediaqueries/
Hi,
Bastien wrote:
> Viktor Rosenfeld <listuser36@googlemail.com> writes:
>
> >> Can you make an alternate .css with non-fixed top bar and a better font?
> >
> > Maybe on the weekend, but I can't promise anything.
>
> Thanks in advance for this!
What's the git URL of the homepage repository?
Thanks,
Viktor
> > What's the git URL of the homepage repository? > git clone git://orgmode.org/orgweb.git -- Eric Schulte http://cs.unm.edu/~eschulte/
[-- Attachment #1: Type: text/plain, Size: 1071 bytes --] >> >> Mh.. your screen seems quite small. If you can fix the .css to display >> the website better on your screen, please do, I don't have time at hand >> now to do it myself. >> The attached three patches to the orgweb repository change the CSS for smarter rendering on narrow screens (they don't address screen height). The changes include... - scale down images on narrow screens - decrease the width of the left link bar on narrow screens - decrease the padding around the title on narrow screens - shrink the paypal link on smaller screens -- I would like to move this button on really small screens, but somehow that doesn't seem possible - remove the Org-mode image on really small screens - remove the twitter feed on narrow screens -- for some reason I was unable to change the size of this widget, so I just hide it on a tiny screen These changes make the new website work on my system, and should improve the reading experience for everyone who keeps their browser screens less than 1400 pixels wide. If these look good please apply them. Thanks, [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #2: 0001-ignoring-exported-html-indices.patch --] [-- Type: text/x-patch, Size: 460 bytes --] From 9bb86f8c206e2b749c9c49c4ad33f11e07e4f3fc Mon Sep 17 00:00:00 2001 From: Eric Schulte <eric.schulte@gmx.com> Date: Wed, 14 Dec 2011 20:30:02 -0700 Subject: [PATCH 1/3] ignoring exported html indices --- .gitignore | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) create mode 100644 .gitignore diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..dcaf716 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +index.html -- 1.7.8 [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #3: 0002-shrink-links-and-remove-twitter-on-narrow-screens.patch --] [-- Type: text/x-patch, Size: 3880 bytes --] From e262903d6d21173d786c15d4e4db155d466c8407 Mon Sep 17 00:00:00 2001 From: Eric Schulte <eric.schulte@gmx.com> Date: Wed, 14 Dec 2011 20:30:28 -0700 Subject: [PATCH 2/3] shrink links and remove twitter on narrow screens --- org.css | 147 +++++++++++++++++++++++++++++++++++---------------------------- 1 files changed, 82 insertions(+), 65 deletions(-) diff --git a/org.css b/org.css index cce24dd..ad6785e 100644 --- a/org.css +++ b/org.css @@ -96,50 +96,87 @@ h1.title { font-family: Courier New; } -#linklist -{ - position: fixed; - font-size: 13pt; - font-family: Courier New; - padding-top: 0px; - padding-right: 0px; - top: 107px; - left: 0px; - margin-top: 0px; - width: 180px; - background-color: #fff; - color: black; - box-shadow: 8px 8px 12px #ccc; - -webkit-border-bottom-right-radius: 10px; - -moz-border-radius-bottomright: 10px; - z-index: 100; -} - -#linklist a { - color: black; - font-weight: normal; - text-decoration: none; - display:block; - padding: 7pt; -} - -#linklist ul { - margin: 0; - padding: 0; -} - -#linklist li { - text-align: right; - margin: 0; -} - -.timestamp { - font-family: Courier New; - color: #888888; -} - -#linklist li:hover { - border-left: 7px solid #537d7b; +@media all { + #linklist + { + position: fixed; + font-size: 13pt; + font-family: Courier New; + padding-top: 0px; + padding-right: 0px; + top: 107px; + left: 0px; + margin-top: 0px; + width: 180px; + background-color: #fff; + color: black; + box-shadow: 8px 8px 12px #ccc; + -webkit-border-bottom-right-radius: 10px; + -moz-border-radius-bottomright: 10px; + z-index: 100; + } + + #linklist a { + color: black; + font-weight: normal; + text-decoration: none; + display:block; + padding: 7pt; + } + + #linklist ul { + margin: 0; + padding: 0; + } + + #linklist li { + text-align: right; + margin: 0; + } + + .timestamp { + font-family: Courier New; + color: #888888; + } + + #linklist li:hover { + border-left: 7px solid #537d7b; + } + + #twit { + /* -moz-opacity:.2; */ + /* opacity: .2; */ + /* filter:alpha(opacity=20); */ + position: fixed; + top: 362px; + box-shadow: 8px 8px 12px #ccc; + -webkit-border-bottom-right-radius: 10px; + -moz-border-radius-bottomright: 10px; + z-index: 100; + } + + .outline-2 { + position: relative; + top: 105px; + left: 215px; + width: 75%; + padding-bottom: 5pt; + } +} + +@media all and (max-width: 700px){ + #linklist{ + width: 130px; + } + #linklist a{ + font-size: 10pt; + } + #twit{ + display: none; + } + .outline-2 { + left: 145px; + } } pre { @@ -176,26 +213,6 @@ pre { filter:alpha(opacity=100); } -.outline-2 { - position: relative; - left: 215px; - top: 105px; - width: 75%; - padding-bottom: 5pt; -} - -#twit { - /* -moz-opacity:.2; */ - /* opacity: .2; */ - /* filter:alpha(opacity=20); */ - position: fixed; - top: 362px; - box-shadow: 8px 8px 12px #ccc; - -webkit-border-bottom-right-radius: 10px; - -moz-border-radius-bottomright: 10px; - z-index: 100; -} - /* #twit:hover { */ /* -moz-opacity:1; */ /* opacity: 1; */ @@ -314,7 +331,7 @@ li { } img.random { - max-width: 750px; + max-width: 75%; max-height: 380px; margin-bottom: 10pt; border: 1px solid black; -- 1.7.8 [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #4: 0003-title-width-adjusts-appropriately-given-window-width.patch --] [-- Type: text/x-patch, Size: 2483 bytes --] From 67762c50047bbbfb71d87245b791c02cc9dcf234 Mon Sep 17 00:00:00 2001 From: Eric Schulte <eric.schulte@gmx.com> Date: Wed, 14 Dec 2011 21:02:28 -0700 Subject: [PATCH 3/3] title width adjusts appropriately given window width --- org.css | 69 ++++++++++++++++++++++++++++++++++++++++++++++---------------- 1 files changed, 51 insertions(+), 18 deletions(-) diff --git a/org.css b/org.css index ad6785e..8cbe0c3 100644 --- a/org.css +++ b/org.css @@ -13,7 +13,6 @@ body { } .title { - background: url(http://orgmode.org/org-mode-unicorn.png) no-repeat 12px 5px; position: fixed; display: inline; left: 0px; @@ -41,14 +40,6 @@ body { z-index: 98; } -#paypal { - position:fixed; - right: 10px; - top: 15px; - z-index: 100; - text-align: center; -} - #paypal a { font-family: Courier new; cursor: pointer; @@ -88,15 +79,23 @@ body { padding-top: 15px; } -h1.title { - text-shadow: 2px 2px 4px #999; - padding-top: 23px; - padding-left: 70pt; - font-size: 23pt; - font-family: Courier New; -} - @media all { + #paypal { + position:fixed; + right: 10px; + top: 15px; + z-index: 100; + text-align: center; + } + + h1.title { + text-shadow: 2px 2px 4px #999; + padding-top: 23px; + padding-left: 0pt; + font-size: 23pt; + font-family: Courier New; + } + #linklist { position: fixed; @@ -164,7 +163,36 @@ h1.title { } } +@media all and (min-width: 1400px){ + padding-left: 75pt; +} + +@media all and (min-width: 650px){ + .title{ + background: url(http://orgmode.org/org-mode-unicorn.png) no-repeat 12px 5px; + } +} + +@media all and (max-width: 750px){ + #paypal{ + position: absolute; + top: 2px; + right: 2px; + } + #paypal a{ + width: 60px; + font-size: 8pt; + } + #paypal a:hover { + width: 60px; + font-size: 8pt; + } +} + @media all and (max-width: 700px){ + h1.title{ + padding-left: 0pt; + } #linklist{ width: 130px; } @@ -174,9 +202,14 @@ h1.title { #twit{ display: none; } - .outline-2 { + .outline-2{ left: 145px; } + #paypal{ + position:fixed; + left: 0px; + top: 362px; + } } pre { -- 1.7.8 [-- Attachment #5: Type: text/plain, Size: 47 bytes --] -- Eric Schulte http://cs.unm.edu/~eschulte/
Bastien <bzg@altern.org> writes:
> Dear all,
>
> I made a small refresh of the website.
>
> http://orgmode.org
>
> - random quote (taken from worg/org-quotes.org)
> - random screenshot
> - twitter feed
> - g+/flattr/fb buttons
> - french translation (http://orgmode.org/fr/)
>
> If you want to help with the translation in your language,
> please send your public key and I will give you push access.
> This is just a bunch of .org files, only 1 hour needed to
> translate the whole website.
>
> Hope you like it!
I personally really like the new look (on a 19" Wide-screen Safari browser under
OSX). It looks really clean, organized, and professional.
I tended to think the original page held just too much information, as a
landing page - i.e. for a potentially new Org-user who is making there
first steps into researching Org-mode.
Incidentally, it also looks really good on the Ipad (safari), where the left-hand
column links work just fine.
Best, Martyn
Hi Eric, Eric Schulte <eric.schulte@gmx.com> writes: > The attached three patches to the orgweb repository change the CSS for > smarter rendering on narrow screens (they don't address screen > height). > > The changes include... > - scale down images on narrow screens > - decrease the width of the left link bar on narrow screens > - decrease the padding around the title on narrow screens > - shrink the paypal link on smaller screens -- I would like to move this > button on really small screens, but somehow that doesn't seem possible > - remove the Org-mode image on really small screens > - remove the twitter feed on narrow screens -- for some reason I was > unable to change the size of this widget, so I just hide it on a tiny > screen Looks fine - thanks! > These changes make the new website work on my system, and should improve > the reading experience for everyone who keeps their browser screens less > than 1400 pixels wide. > > If these look good please apply them. I don't understand the first patch, as there should be no index.html in the git repository. The two others don't apply. There seem to be blank lines where there should not. Can you resent them? Thanks! -- Bastien
Hi Eric, Eric Schulte <eric.schulte@gmx.com> writes: > Ah, I should have known to hold my tongue! :) > > I pushed a patch up to a new "old-image" branch in the orgweb > repository, but it seems you've now taken care of this independently so > please feel free to delete that branch. Mhh.. yes, I deleted this branch. Thanks anyway! > One of my preferred layout looks something like the following, which can > lead to a small windows for Firefox. > > +----------------+---------------+ > | | | > | | | > | Firefox | Emacs | > | | | > | | | > +----------------+---------------+ > | Terminal | > +--------------------------------+ > > There is a huge difference in readability between those websites which > take variable window size into consideration and those which assume a > large screen (or a full screen browser). I'll take a shot at changing > the CSS in the orgweb repository to facility smaller layouts (I guess > some special CSS for mobile browsers may also make sense). > > ...looking... > > It looks like making the Org-mode website responsive to the viewers > window size should be as simple as adding a couple of "@media" guards > [1] to the css page. I may have time to tackle this over the weekend. > I do need some help compiling the Org-mode web page from the git > repository. Is there an org-mode publishing project which I should > define locally? I use this locally: #+begin_src emacs-lisp (setq org-publish-project-alist `(("orgweb" :base-directory "~/install/git/orgweb/" :base-extension "org" :publishing-directory "/srv/http/org-mode/" :publishing-function org-publish-org-to-html :auto-sitemap nil :section-numbers nil :table-of-contents nil :html-preamble ,(org-get-file-contents "/srv/http/org-mode/preamble.html") :html-postamble ,(org-get-file-contents "/srv/http/org-mode/postamble.html")) ("orgweb_extra" :base-directory "~/install/git/orgweb/" :base-extension "css\\|html\\|png" :publishing-directory "/srv/http/org-mode/" :publishing-function org-publish-attachment))) #+end_src ... my web server serves web pages at "/srv/http/org-mode/". I also have a project for the fr/ directory, very similar to this one. > In general how are the many file in orgweb compiled into the site? Each time someone pushes a fix to orgweb.git, there is a hook calling a script launching emacs in batch mode for publishing the website. The same way than for Worg, exactly. HTH, -- Bastien
Bastien <bzg@altern.org> writes: > Hi Eric, > > Eric Schulte <eric.schulte@gmx.com> writes: > >> The attached three patches to the orgweb repository change the CSS for >> smarter rendering on narrow screens (they don't address screen >> height). >> >> The changes include... >> - scale down images on narrow screens >> - decrease the width of the left link bar on narrow screens >> - decrease the padding around the title on narrow screens >> - shrink the paypal link on smaller screens -- I would like to move this >> button on really small screens, but somehow that doesn't seem possible >> - remove the Org-mode image on really small screens >> - remove the twitter feed on narrow screens -- for some reason I was >> unable to change the size of this widget, so I just hide it on a tiny >> screen > > Looks fine - thanks! > >> These changes make the new website work on my system, and should improve >> the reading experience for everyone who keeps their browser screens less >> than 1400 pixels wide. >> >> If these look good please apply them. > > I don't understand the first patch, as there should be no index.html in > the git repository. > Exactly, that is why it is ignored. :) I did this mainly so that I could export index.org to generate a base to work off of (before I had the project description you just send in another email). Should the publish project definition be added to the orgweb repository in a non-exporting file, so that anyone can build the site locally for testing? > > The two others don't apply. There seem to be blank lines where there > should not. Can you resent them? > Since you seem to like the changes listed above I've just pushed these patches up directly, so they should now be applied. Best -- Eric > > Thanks! -- Eric Schulte http://cs.unm.edu/~eschulte/
Hi Eric, Eric Schulte <eric.schulte@gmx.com> writes: > Exactly, that is why it is ignored. :) > > I did this mainly so that I could export index.org to generate a base to > work off of (before I had the project description you just send in > another email). Okay, no problem. > Should the publish project definition be added to the orgweb repository > in a non-exporting file, so that anyone can build the site locally for > testing? Sure -- maybe just add a README. >> The two others don't apply. There seem to be blank lines where there >> should not. Can you resent them? > > Since you seem to like the changes listed above I've just pushed these > patches up directly, so they should now be applied. Thanks. We lost the blue background in the top bar and the correct z-index so that text don't overlap. -- Bastien
> > Thanks. We lost the blue background in the top bar and the correct > z-index so that text don't overlap. I believe I just pushed up a fix for this -- Eric Schulte http://cs.unm.edu/~eschulte/
Eric Schulte <eric.schulte@gmx.com> writes:
>> Thanks. We lost the blue background in the top bar and the correct
>> z-index so that text don't overlap.
>
> I believe I just pushed up a fix for this
Confirmed, thanks!
--
Bastien
Hi Achim,
Achim Gratz <Stromeko@nexgo.de> writes:
> The wayback archive has a correct version of that file:
> http://web.archive.org/web/20100925104250/http://orgmode.org/org-mode-unicorn.ico
Thanks -- both org-mode-unicorn.ico and favicon.ico should work
fine now.
--
Bastien
Hi, I can't get SBE blocks called from a table to work, if they use functions like org-parse-time-string. Suppose I have the following table: | Start | Ende | | |------------------+------------------+-| | [2011-06-29 Wed] | [2012-02-29 Wed] | | #+TBLFM: $3='(sbe "billable-month" (start $1) (end $2)) and the following source block: #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]") #+BEGIN_SRC emacs-lisp (let* ((start-date start)) (message "%s" start-date)) #+END_SRC If I evaluate the table, the start date is put into the last column. However, if I change the code to the following: #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]") #+BEGIN_SRC emacs-lisp (let* ((start-date (org-parse-time-string start))) (message "%s" (nth 4 start-date))) #+END_SRC then the string #ERROR is inserted into the table. Evaluating the source block directly yields the correct result. What's going on here? Thanks, Viktor
Viktor Rosenfeld <listuser36@googlemail.com> wrote:
> Hi,
>
> I can't get SBE blocks called from a table to work, if they use
> functions like org-parse-time-string.
>
> Suppose I have the following table:
>
> | Start | Ende | |
> |------------------+------------------+-|
> | [2011-06-29 Wed] | [2012-02-29 Wed] | |
> #+TBLFM: $3='(sbe "billable-month" (start $1) (end $2))
>
> and the following source block:
>
> #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]")
> #+BEGIN_SRC emacs-lisp
> (let* ((start-date start))
> (message "%s" start-date))
> #+END_SRC
>
> If I evaluate the table, the start date is put into the last column.
> However, if I change the code to the following:
>
> #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]")
> #+BEGIN_SRC emacs-lisp
> (let* ((start-date (org-parse-time-string start)))
> (message "%s" (nth 4 start-date)))
> #+END_SRC
>
> then the string #ERROR is inserted into the table. Evaluating the source
> block directly yields the correct result.
>
> What's going on here?
>
Finicky type matching: if you evaluate the second code block in the buffer with
ESC ESC : (sbe "billable-month" (start "[2011-06-29 Wed]") (end "[2012-02-29 Wed]")) RET
you get a backtrace similar to this:
,----
| Debugger entered--Lisp error: (wrong-type-argument stringp [2011-06-29 Wed])
| string-match("\\(\\([0-9]\\{4\\}\\)-\\([0-9]\\{2\\}\\)-\\([0-9]\\{2\\}\\)\\( +[^]+0-9>
\n -]+\\)?\\( +\\([0-9]\\{1,2\\}\\):\\([0-9]\\{2\\}\\)\\)?\\)" [2011-06-29 Wed])
| (if (string-match org-ts-regexp0 s) (list 0 (if (or (match-beginning 8) (not nodefault)) (string-to-number (or (match-string 8 s) "0"))) (if (or (match-beginning 7) (not nodefault)) (string-to-number (or (match-string 7 s) "0"))) (string-to-number (match-string 4 s)) (string-to-number (match-string 3 s)) (string-to-number (match-string 2 s)) nil nil nil) (error "Not a standard Org-mode time string: %s" s))
| org-parse-time-string([2011-06-29 Wed])
| (let* ((start-date (org-parse-time-string start))) (format "%d" (nth 4 start-date)))
| ...
`----
Somewhere, the string becomes not a string. Try modifying the block to this:
#+BEGIN_SRC emacs-lisp
(let* ((start-date (org-parse-time-string (format "%s" start))))
(message "%s" (nth 4 start-date)))
#+END_SRC
Nick
Hi Nick,
Nick Dokos wrote:
> Viktor Rosenfeld <listuser36@googlemail.com> wrote:
>
> > Hi,
> >
> > I can't get SBE blocks called from a table to work, if they use
> > functions like org-parse-time-string.
> >
> > Suppose I have the following table:
> >
> > | Start | Ende | |
> > |------------------+------------------+-|
> > | [2011-06-29 Wed] | [2012-02-29 Wed] | |
> > #+TBLFM: $3='(sbe "billable-month" (start $1) (end $2))
> >
> > and the following source block:
> >
> > #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]")
> > #+BEGIN_SRC emacs-lisp
> > (let* ((start-date start))
> > (message "%s" start-date))
> > #+END_SRC
> >
> > If I evaluate the table, the start date is put into the last column.
> > However, if I change the code to the following:
> >
> > #+NAME: billable-month(start="[2011-06-29 Wed]", end="[2012-02-29 Wed]")
> > #+BEGIN_SRC emacs-lisp
> > (let* ((start-date (org-parse-time-string start)))
> > (message "%s" (nth 4 start-date)))
> > #+END_SRC
> >
> > then the string #ERROR is inserted into the table. Evaluating the source
> > block directly yields the correct result.
> >
> > What's going on here?
> >
>
> Finicky type matching: if you evaluate the second code block in the buffer with
>
> ESC ESC : (sbe "billable-month" (start "[2011-06-29 Wed]") (end "[2012-02-29 Wed]")) RET
>
> you get a backtrace similar to this:
>
> ,----
> | Debugger entered--Lisp error: (wrong-type-argument stringp [2011-06-29 Wed])
> | string-match("\\(\\([0-9]\\{4\\}\\)-\\([0-9]\\{2\\}\\)-\\([0-9]\\{2\\}\\)\\( +[^]+0-9>
> \n -]+\\)?\\( +\\([0-9]\\{1,2\\}\\):\\([0-9]\\{2\\}\\)\\)?\\)" [2011-06-29 Wed])
> | (if (string-match org-ts-regexp0 s) (list 0 (if (or (match-beginning 8) (not nodefault)) (string-to-number (or (match-string 8 s) "0"))) (if (or (match-beginning 7) (not nodefault)) (string-to-number (or (match-string 7 s) "0"))) (string-to-number (match-string 4 s)) (string-to-number (match-string 3 s)) (string-to-number (match-string 2 s)) nil nil nil) (error "Not a standard Org-mode time string: %s" s))
> | org-parse-time-string([2011-06-29 Wed])
> | (let* ((start-date (org-parse-time-string start))) (format "%d" (nth 4 start-date)))
> | ...
> `----
>
> Somewhere, the string becomes not a string. Try modifying the block to this:
>
> #+BEGIN_SRC emacs-lisp
> (let* ((start-date (org-parse-time-string (format "%s" start))))
> (message "%s" (nth 4 start-date)))
> #+END_SRC
>
> Nick
Thanks, that does the trick. The double ESC evaluation is helpful, too.
Cheers,
Viktor