Angus Leeming <[EMAIL PROTECTED]> writes:
| Given that these functions are all declared in a single header file,
| lyxlib.h, is there a reason why each single function is defined in its own
| .C file?
Yes, keeping things super simple.
| abort.C atoi.C chdir.C copy.C getcwd.C kill.C mkdir.C put
On Thu, Aug 28, 2003 at 07:00:42PM +, Angus Leeming spake thusly:
> I've been sitting on this all month, but I do think that this is about ready
> to go in.
>
> Attached is a rewrite of the BibTeX inset so that it derives direct from
> InsetOld rather than from the totally inappropriate In
On Friday 05 September 2003 8:36 am, Martin Vermeer wrote:
> > Attached is a rewrite of the BibTeX inset so that it derives direct from
> > InsetOld rather than from the totally inappropriate InsetCommand.
> Whatever became of this patch?
The FileName class can't handle the case of the file being
Michael Schmitt wrote:
>>Log message:
>>Strip out redundant #includes
> Angus, how did you determine all theses redundant #includes? Is there
> tools available that detects them automatically?
http://article.gmane.org/gmane.editors.lyx.devel:27547
--
Angus
Hi,
1. if you change your spellchecking language to aspell and try to find your
personal dictionary, the default for file type is still "*.ispell", and it
cannot be changed. aspell personal dictionaries are normally something like
".aspell..". Even my file .ispell_german is not displayed
2.
Angus Leeming <[EMAIL PROTECTED]> writes:
| On Friday 05 September 2003 8:36 am, Martin Vermeer wrote:
>> > Attached is a rewrite of the BibTeX inset so that it derives direct from
>> > InsetOld rather than from the totally inappropriate InsetCommand.
>
>> Whatever became of this patch?
>
| The Fi
Angus Leeming <[EMAIL PROTECTED]> writes:
| Michael Schmitt wrote:
>>>Log message:
>>>Strip out redundant #includes
>
>> Angus, how did you determine all theses redundant #includes? Is there
>> tools available that detects them automatically?
>
| http://article.gmane.org/gmane.editors.lyx.devel:27
On Fri, Sep 05, 2003 at 01:44:18AM +, Angus Leeming wrote:
> If I were to concatenate them together, including header blurb, the resuling
> file would be 363 lines long. Hardly enormous...
All massive header files start small. inset.h was once 140 lines long.
> Any objections to me doing th
On Thu, Sep 04, 2003 at 10:40:16AM +0200, Hartmut Haase wrote:
> 1. if you change your spellchecking language to aspell and try to find your
> personal dictionary, the default for file type is still "*.ispell", and it
> cannot be changed. aspell personal dictionaries are normally something like
John Levon wrote:
> On Fri, Sep 05, 2003 at 01:44:18AM +, Angus Leeming wrote:
>
>> If I were to concatenate them together, including header blurb, the
resuling
>> file would be 363 lines long. Hardly enormous...
>
> All massive header files start small. inset.h was once 140 lines long.
>
On Fri, Sep 05, 2003 at 10:22:46AM +, Angus Leeming wrote:
> I don't think you slept well last night ;-)
I did actually ... haven't quite woken up yet though. A brisk cup of
tea is needed I think.
> Your comments don't make much
> sense to me. (You seem to be objecting to me concatenating
Lars Gullik Bjønnes wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
>
> | On Friday 05 September 2003 8:36 am, Martin Vermeer wrote:
>>> > Attached is a rewrite of the BibTeX inset so that it derives direct from
>>> > InsetOld rather than from the totally inappropriate InsetCommand.
>>
>>> Wha
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>
>> Angus Leeming <[EMAIL PROTECTED]> writes:
>>
>> | On Friday 05 September 2003 8:36 am, Martin Vermeer wrote:
> Attached is a rewrite of the BibTeX inset so that it derives direct from
> InsetOld rather than from
Lars Gullik Bjønnes wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
> | Yup, your SOH is restored ;-)
> SOH?
Sense Of Humour...
--
Angus
Having tried to understand what goes on in text3.C's dispatch on
LFUN_MOUSE_RELEASE. Those methods checkInsetHit() and checkInset()...
would it be possible to simplify this by adding a method hit(x,y) to
insets?
Here, x,y would be *absolute* co-ordinates (Insets "know" their own
origin's absolut
Hello, Jean-Marc
On Thu, 4 Sep 2003, Jean-Marc Lasgouttes wrote:
> > "Joao" == Joao Luis Meloni Assirati <[EMAIL PROTECTED]> writes:
>
> Joao> Therefore, I suggest that when '\command{blah blah}' is unknown
> Joao> to tex2lyx, it be translated as a single ERT. Maybe to add a
> Joao> command
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>
>> Angus Leeming <[EMAIL PROTECTED]> writes:
>> | Yup, your SOH is restored ;-)
>> SOH?
>
| Sense Of Humour...
are you hitting on me?
--
Lgb
Lars Gullik Bjønnes wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
>
> | Lars Gullik Bjønnes wrote:
>>
>>> Angus Leeming <[EMAIL PROTECTED]> writes:
>>> | Yup, your SOH is restored ;-)
>>> SOH?
>>
> | Sense Of Humour...
>
> are you hitting on me?
Merely refering to your comment earlier abou
Angus Leeming <[EMAIL PROTECTED]> writes:
| Lars Gullik Bjønnes wrote:
>
>> Angus Leeming <[EMAIL PROTECTED]> writes:
>>
>> | Lars Gullik Bjønnes wrote:
>>>
Angus Leeming <[EMAIL PROTECTED]> writes:
| Yup, your SOH is restored ;-)
SOH?
>>>
>> | Sense Of Humour...
>>
>> are you hit
Lars Gullik Bjønnes wrote:
> PPS. Why do you thing that "gayness" is visible?
This mailing list goes from strength to strength!
--
Angus
On Fri, Sep 05, 2003 at 12:23:42PM +, Angus Leeming wrote:
> > PPS. Why do you thing that "gayness" is visible?
>
> This mailing list goes from strength to strength!
It is Friday !
john
--
Khendon's Law:
If the same point is made twice by the same person, the thread is over.
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> Michael Schmitt wrote:
>>> Log message: Strip out redundant #includes
>> Angus, how did you determine all theses redundant #includes? Is
>> there tools available that detects them automatically?
Angus> http://article.gmane.org/gma
Jean-Marc Lasgouttes <[EMAIL PROTECTED]> writes:
| So maybe a less aggressive approach in your scripts could be to remove
| a header, check that it still compiles _and_ that the header does not
| show in the dependencies anymore. This will allow to remove really
| useless headers.
I think the agg
On Friday 05 September 2003 12:01 pm, Jean-Marc Lasgouttes wrote:
> Angus> http://article.gmane.org/gmane.editors.lyx.devel:27547
>
> What I do not necessarily like about this approach is that it does not
> remove #includes from _headers_ which is the most reasonable thing to
> do. Having extra inc
On Fri, Sep 05, 2003 at 01:22:26PM +0300, Martin Vermeer spake thusly:
>
> Having tried to understand what goes on in text3.C's dispatch on
> LFUN_MOUSE_RELEASE. Those methods checkInsetHit() and checkInset()...
> would it be possible to simplify this by adding a method hit(x,y) to
> insets?
>
> "Angus" == Angus Leeming <[EMAIL PROTECTED]> writes:
Angus> I agree with you. However, I think of this as an iterative
Angus> process. As Lars suggests, the next step is to do the same for
Angus> the header files themselves.
Maybe one thing that would be worth checking is how much you can
r
On Fri, Sep 05, 2003 at 03:51:38AM +, Angus Leeming spake thusly:
>
> Angus
I am afraid that is needed in math_extern.C.
For find_if().
- Martin
pgp0.pgp
Description: PGP signature
On Friday 05 September 2003 12:42 pm, Jean-Marc Lasgouttes wrote:
> Maybe one thing that would be worth checking is how much you can
> remove from a header while keeping it self contained, so that
> #include
> #include "foo.h"
> still compiles.
Good idea actually.
> Can you tell what you gai
On Fri, Sep 05, 2003 at 03:51:38AM +, Angus Leeming spake thusly:
> I thought I'd see if I could strip out the redundant #includes,
> so wrote a little script to remove each one in turn from a file
> and ascertain whether the thing would still compile. It's a bit
> slow but the results for
Lars I remember this was your baby though I cannot find the posting.
You need the attached patch to compile... is this correct?
- Martin
Index: lstrings.C
===
RCS file: /usr/local/lyx/cvsroot/lyx-devel/src/support/lstrings.C,v
retr
Martin Vermeer <[EMAIL PROTECTED]> writes:
| Lars I remember this was your baby though I cannot find the posting.
no, bformat is Andre's baby.
--
Lgb
Martin Vermeer <[EMAIL PROTECTED]> writes:
| Lars I remember this was your baby though I cannot find the posting.
>
| You need the attached patch to compile... is this correct?
Yes, I guess that would work.
--
Lgb
support/textutils.h cannot currently be compiled stand-alone. To do so would
require a
#include "paragraph.h".
inline
bool IsInsetChar(char c)
{
return (c == Paragraph::META_INSET);
}
So, should I do that or should I move this function out-of-line?
$ grep -r 'textutils.h' .
./buffer.C:#
My take on ChangeLogs is this:
They are a complete waste of time, and only bloats the distribution.
Write what you did in the commit log for CVS, and use view CVS to browse
the code when you are looking for regressions.
Best regards,
Asger
On Fri, 5 Sep 2003, Angus Leeming wrote:
> support/textutils.h cannot currently be compiled stand-alone. To do so would
> require a
> #include "paragraph.h".
>
> inline
> bool IsInsetChar(char c)
> {
> return (c == Paragraph::META_INSET);
> }
>
> So, should I do that or should I move this
On Fri, Sep 05, 2003 at 05:53:41PM +0200, Asger Kunuk Alstrup wrote:
> Write what you did in the commit log for CVS, and use view CVS to browse
> the code when you are looking for regressions.
CVS can only handle per-file changesets. It's practically useless for
that sort of thing.
regards
john
I have finished making huge changes to the tree, removing redundant #includes
and renaming LString.h, Lsstream.h, LIstream.h and LOStream.h. FWIW here's my
tally of numbers of #includes removed from the source:
mathed 168
graphics26
insets 193
src 320
fronte
On Fri, Sep 05, 2003 at 03:51:38AM +, Angus Leeming spake thusly:
>
> I thought I'd see if I could strip out the redundant #includes,
> so wrote a little script to remove each one in turn from a file
> and ascertain whether the thing would still compile. It's a bit
> slow but the results fo
On Friday 05 September 2003 7:02 pm, Martin Vermeer wrote:
> in biblio.C, frnt_lang.C, xfont_loader.C, graph.C,
> texrow.C...
> in Alert_pimpl.h, ...
> in Color.C, ColorHandler.C, ...
> in FontInfo.C...
> in BranchList.C...
> in factory.Ci, vspace.C
> (because of sscanf... isn't that pl
On Fri, Sep 05, 2003 at 07:47:34PM +, Angus Leeming spake thusly:
> On Friday 05 September 2003 7:02 pm, Martin Vermeer wrote:
> > in biblio.C, frnt_lang.C, xfont_loader.C, graph.C,
> > texrow.C...
> > in Alert_pimpl.h, ...
> > in Color.C, ColorHandler.C, ...
> > in FontInfo.C...
> > in
I wrote a couple of scripts to extract some numbers from the dependency files
to be found in the .deps dirs. See attached.
$ ./deps.sh build/src/insets/ExternalTemplates.o
ExternalTemplate.o depends on 151 system files and 31 local files
If you pass deps.sh all the .o files in the build, dumping
Martin Vermeer wrote:
>> Could I get you to just commit a patch?
>
> Yes, here's the patch. Note that I didn't look very carefully at where
> I put the headers, I just wanted to get it to compile. So you may have
> a little post-cleanup to do. Please you apply it as I have some other
> small stuff
Lars Gullik Bjønnes wrote:
> Angus Leeming <[EMAIL PROTECTED]> writes:
>
> | Given that these functions are all declared in a single header file,
> | lyxlib.h, is there a reason why each single function is defined in its own
> | .C file?
>
> Yes, keeping things super simple.
>
> | abort.C atoi.
Martin Vermeer wrote:
>
> OK... but this kind of thing should be done by someone with an old and
> finicky compiler. You're still in for some qt flak BTW and I wouldn't
> be surprised if the minority platform people (Rob?) will release some
> profanities.
I am not actively using LyX CVS now. CV
On Sat, Sep 06, 2003 at 09:00:55AM +0900, Rob Lahaye spake thusly:
>
> Martin Vermeer wrote:
> >
> > OK... but this kind of thing should be done by someone with an old and
> > finicky compiler. You're still in for some qt flak BTW and I wouldn't
> > be surprised if the minority platform people
On Fri, Sep 05, 2003 at 11:18:18PM +, Angus Leeming spake thusly:
> Martin Vermeer wrote:
> >> Could I get you to just commit a patch?
> >
> > Yes, here's the patch. Note that I didn't look very carefully at where
> > I put the headers, I just wanted to get it to compile. So you may have
> >
46 matches
Mail list logo
