Re: Mathed delimiters bug
Andre == Andre Poenitz [EMAIL PROTECTED] writes: Andre It should be ( ) which gets parsed as ls = ( and rs = ) Andre and serves as input for the delim inset. OK, I did something like that. BTW Andre', did you ever notice that the left and right delimiters are not at the same height on screen? It seems there is a one pixel shift. JMarc
Re: Mathed delimiters bug
On Tue, Apr 16, 2002 at 12:00:19PM +0200, Jean-Marc Lasgouttes wrote: OK, I did something like that. BTW Andre', did you ever notice that the left and right delimiters are not at the same height on screen? It seems there is a one pixel shift. Yes. I guess that's because the right one is a rotated version of the left one and some rounding comes into play. But this has always been there. Does this need a fix? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
Andre == Andre Poenitz [EMAIL PROTECTED] writes: Andre Yes. I guess that's because the right one is a rotated version Andre of the left one and some rounding comes into play. But this has Andre always been there. Does this need a fix? I filed bug #327 for it, but I agree it is just cosmetic. JMarc
Re: Mathed delimiters bug
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes:
Andre> It should be "( )" which gets parsed as ls = "(" and rs = ")"
Andre> and serves as input for the delim inset.
OK, I did something like that. BTW Andre', did you ever notice that
the left and right delimiters are not at the same height on screen? It
seems there is a one pixel shift.
JMarc
Re: Mathed delimiters bug
On Tue, Apr 16, 2002 at 12:00:19PM +0200, Jean-Marc Lasgouttes wrote: > OK, I did something like that. BTW Andre', did you ever notice that > the left and right delimiters are not at the same height on screen? It > seems there is a one pixel shift. Yes. I guess that's because the right one is a rotated version of the left one and some rounding comes into play. But this has "always" been there. Does this need a fix? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> Yes. I guess that's because the right one is a rotated version Andre> of the left one and some rounding comes into play. But this has Andre> "always" been there. Does this need a fix? I filed bug #327 for it, but I agree it is just cosmetic. JMarc
Re: Mathed delimiters bug
Andre == Andre Poenitz [EMAIL PROTECTED] writes: Andre On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes Andre wrote: When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' Andre I see the first line, but not the second. And it inserts () and Andre []. I just removed the spurious message. I still fails for me, probably because I am using lyxstring and either (1) our use of istringstream is wrong or (2) there is a bug in lyxstring. However, the following simple patch just works, as far as I know. JMarc patch-delim Description: Binary data
Re: Mathed delimiters bug
-BEGIN PGP SIGNED MESSAGE- On Monday, 15. April 2002 17:09, Jean-Marc Lasgouttes wrote: Andre On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes Andre wrote: When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' Andre I see the first line, but not the second. And it inserts () and Andre []. I just removed the spurious message. I still fails for me, probably because I am using lyxstring and either (1) our use of istringstream is wrong or (2) there is a bug in lyxstring. However, the following simple patch just works, as far as I know. Works for me too. (I mean, now the math-Dialog with parenthesis, which was not working for me before) Kornel - -- Kornel Benko [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: PGP 6.5.8 iQCVAwUBPLrvNrewfbDGmeqhAQFBcQQA1QJsNuIuM7cAGRCyAMbV1Icir2mNjykM cJoXY4fgYbvzx11uI5uGMHw89rHTT6h33czlkJbnYqCHI4TIbKuGKP7nhu8sRWac anMZ/+px3wnt/j/rEpmlzhyIvxZkBQFJ4Z0MP0zdh1/3kyIicK8PY4KkkhTviySy lIacK1QgVvY= =ix+c -END PGP SIGNATURE-
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:09:30PM +0200, Jean-Marc Lasgouttes wrote: I still fails for me, probably because I am using lyxstring and either (1) our use of istringstream is wrong I bet a sixpack of beer on that. or (2) there is a bug in lyxstring. Actually I rather think the istringstreams are broken. But I would bet just a single bottle on that. However, the following simple patch just works, as far as I know. So feel free to apply it. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: Works for me too. (I mean, now the math-Dialog with parenthesis, which was not working for me before) Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my local tree that makes committing formulabase.C a bit difficult. Andre' PS: Not that we use istringstream a second time in that file and probably in a few more cases... -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
-BEGIN PGP SIGNED MESSAGE- On Monday, 15. April 2002 17:53, Andre Poenitz wrote: On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: Works for me too. (I mean, now the math-Dialog with parenthesis, which was not working for me before) Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my local tree that makes committing formulabase.C a bit difficult. So, please Jean-Marc, would you like to commit it? The math-delimiter-dialog was broken, and your patch fitts exactly. Kornel - -- Kornel Benko [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: PGP 6.5.8 iQCVAwUBPLr5P7ewfbDGmeqhAQFv9gP8CKk3Qs78mzYekDTVzyESYkTqoeA9Mz0l o3jLq8x9OkGF9c1ja3JCU1NIdJJ70C6amqT6frTUukixE/YiV14GAE+CmA4hb6Ie FDCzH8PdVMErmvGp47O1sq+/ZToqUeOlUVa325eeekQbfd5p6DgGShQ9PjupInJa VXNW/jRGv3s= =cwpr -END PGP SIGNATURE-
Re: Mathed delimiters bug
On Monday 15 April 2002 4:53 pm, Andre Poenitz wrote: On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: Works for me too. (I mean, now the math-Dialog with parenthesis, which was not working for me before) Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my local tree that makes committing formulabase.C a bit difficult. I think you mean bug here, not bugger. That would be a rather extreme way of convincing a happy father to do anything. Angus
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:05:31PM +0100, Angus Leeming wrote: I think you mean bug here, not bugger. That would be a rather extreme way of convincing a happy father to do anything. I guess you are right. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 06:20:29PM +0200, Jean-Marc Lasgouttes wrote: Andre PS: Not that we use istringstream a second time in that file Andre and probably in a few more cases... I've seen that. Do you think I should adapt this code too? I think these parts are never used, but I am not sure. Actually, it would be better if we could convince ourselves that this is the result of some broken istringstream implementation... Does the behaviour change in your case if you do not use 'arg.c_str()' as input for the istringstream, but something like arg + '\0' or arg + ' '? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
Andre == Andre Poenitz [EMAIL PROTECTED] writes: Andre On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes Andre wrote: When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' Andre I see the first line, but not the second. And it inserts () and Andre []. I just removed the spurious message. What did you change to the code? What is the meaning of this () stuff? Shall I keep it? Or set arg to '( )' if it is empty? JMarc
Re: Mathed delimiters bug
Andre Poenitz [EMAIL PROTECTED] writes: | On Mon, Apr 15, 2002 at 05:09:30PM +0200, Jean-Marc Lasgouttes wrote: I still fails for me, probably because I am using lyxstring and either (1) our use of istringstream is wrong | I bet a sixpack of beer on that. or (2) there is a bug in lyxstring. | Actually I rather think the istringstreams are broken. But I would bet just | a single bottle on that. There are bugs in the gcc 2.* versions of libstc++, but I have only encountered them then using a stringstream both as input and output at the same time. -- Lgb
Re: Mathed delimiters bug
Andre Poenitz [EMAIL PROTECTED] writes: | I think we could use | string ls = (; | string rs = ); | instead of | string ls; | string rs; and I think we should use: string ls((); string rs()); but that is just me I guess... -- Lgb
Re: Mathed delimiters bug
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes Andre> wrote: >> When trying to use M-m [ or M-m (, I get respectively >> formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters >> from '[ ]' Andre> I see the first line, but not the second. And it inserts () and Andre> []. I just removed the spurious message. I still fails for me, probably because I am using lyxstring and either (1) our use of istringstream is wrong or (2) there is a bug in lyxstring. However, the following simple patch just works, as far as I know. JMarc patch-delim Description: Binary data
Re: Mathed delimiters bug
-BEGIN PGP SIGNED MESSAGE- On Monday, 15. April 2002 17:09, Jean-Marc Lasgouttes wrote: > Andre> On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes > > Andre> wrote: > >> When trying to use M-m [ or M-m (, I get respectively > >> formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters > >> from '[ ]' > > Andre> I see the first line, but not the second. And it inserts () and > Andre> []. I just removed the spurious message. > > I still fails for me, probably because I am using lyxstring and either > (1) our use of istringstream is wrong or (2) there is a bug in > lyxstring. > > However, the following simple patch just works, as far as I know. Works for me too. (I mean, now the math-Dialog with parenthesis, which was not working for me before) Kornel - -- Kornel Benko [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: PGP 6.5.8 iQCVAwUBPLrvNrewfbDGmeqhAQFBcQQA1QJsNuIuM7cAGRCyAMbV1Icir2mNjykM cJoXY4fgYbvzx11uI5uGMHw89rHTT6h33czlkJbnYqCHI4TIbKuGKP7nhu8sRWac anMZ/+px3wnt/j/rEpmlzhyIvxZkBQFJ4Z0MP0zdh1/3kyIicK8PY4KkkhTviySy lIacK1QgVvY= =ix+c -END PGP SIGNATURE-
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:09:30PM +0200, Jean-Marc Lasgouttes wrote: > I still fails for me, probably because I am using lyxstring and either > (1) our use of istringstream is wrong I bet a sixpack of beer on that. > or (2) there is a bug in lyxstring. Actually I rather think the istringstreams are broken. But I would bet just a single bottle on that. > However, the following simple patch just works, as far as I know. So feel free to apply it. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: > Works for me too. (I mean, now the math-Dialog with parenthesis, which > was not working for me before) Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my local tree that makes committing formulabase.C a bit difficult. Andre' PS: Not that we use istringstream a second time in that file and probably in a few more cases... -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
-BEGIN PGP SIGNED MESSAGE- On Monday, 15. April 2002 17:53, Andre Poenitz wrote: > On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: > > Works for me too. (I mean, now the math-Dialog with parenthesis, which > > was not working for me before) > > Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my > local tree that makes committing formulabase.C a bit difficult. So, please Jean-Marc, would you like to commit it? The math-delimiter-dialog was broken, and your patch fitts exactly. Kornel - -- Kornel Benko [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: PGP 6.5.8 iQCVAwUBPLr5P7ewfbDGmeqhAQFv9gP8CKk3Qs78mzYekDTVzyESYkTqoeA9Mz0l o3jLq8x9OkGF9c1ja3JCU1NIdJJ70C6amqT6frTUukixE/YiV14GAE+CmA4hb6Ie FDCzH8PdVMErmvGp47O1sq+/ZToqUeOlUVa325eeekQbfd5p6DgGShQ9PjupInJa VXNW/jRGv3s= =cwpr -END PGP SIGNATURE-
Re: Mathed delimiters bug
On Monday 15 April 2002 4:53 pm, Andre Poenitz wrote: > On Mon, Apr 15, 2002 at 05:30:42PM +0200, Kornel Benko wrote: > > Works for me too. (I mean, now the math-Dialog with parenthesis, which > > was not working for me before) > > Ok, then bugger Jean-Marc to commit it, I have currently the xo/yo in my > local tree that makes committing formulabase.C a bit difficult. I think you mean "bug" here, not "bugger". That would be a rather extreme way of convincing a happy father to do anything. Angus
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 05:05:31PM +0100, Angus Leeming wrote: > I think you mean "bug" here, not "bugger". That would be a rather extreme > way of convincing a happy father to do anything. I guess you are right. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
On Mon, Apr 15, 2002 at 06:20:29PM +0200, Jean-Marc Lasgouttes wrote: > Andre> PS: Not that we use istringstream a second time in that file > Andre> and probably in a few more cases... > > I've seen that. Do you think I should adapt this code too? I think these parts are never used, but I am not sure. Actually, it would be better if we could convince ourselves that this is the result of some broken istringstream implementation... Does the behaviour change in your case if you do not use 'arg.c_str()' as input for the istringstream, but something like arg + '\0' or arg + ' '? Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Re: Mathed delimiters bug
> "Andre" == Andre Poenitz <[EMAIL PROTECTED]> writes: Andre> On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes Andre> wrote: >> When trying to use M-m [ or M-m (, I get respectively >> formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters >> from '[ ]' Andre> I see the first line, but not the second. And it inserts () and Andre> []. I just removed the spurious message. What did you change to the code? What is the meaning of this () stuff? Shall I keep it? Or set arg to '( )' if it is empty? JMarc
Re: Mathed delimiters bug
Andre Poenitz <[EMAIL PROTECTED]> writes: | On Mon, Apr 15, 2002 at 05:09:30PM +0200, Jean-Marc Lasgouttes wrote: >> I still fails for me, probably because I am using lyxstring and either >> (1) our use of istringstream is wrong > | I bet a sixpack of beer on that. > >> or (2) there is a bug in lyxstring. > | Actually I rather think the istringstreams are broken. But I would bet just | a single bottle on that. There are bugs in the gcc 2.* versions of libstc++, but I have only encountered them then using a stringstream both as input and output at the same time. -- Lgb
Re: Mathed delimiters bug
Andre Poenitz <[EMAIL PROTECTED]> writes:
| I think we could use
>
| string ls = "(";
| string rs = ")";
>
| instead of
>
| string ls;
| string rs;
and I think we should use:
string ls("(");
string rs(")");
but that is just me I guess...
--
Lgb
Mathed delimiters bug
When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' formulabase::LFUN_MATH_DELIM, arg: '( )' can't parse delimeters from '( )' This can't be right. JMarc PS: and delimeters should be delimiters
Re: Mathed delimiters bug
On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes wrote: When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' I see the first line, but not the second. And it inserts () and []. I just removed the spurious message. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
Mathed delimiters bug
When trying to use M-m [ or M-m (, I get respectively formulabase::LFUN_MATH_DELIM, arg: '[ ]' can't parse delimeters from '[ ]' formulabase::LFUN_MATH_DELIM, arg: '( )' can't parse delimeters from '( )' This can't be right. JMarc PS: and "delimeters" should be "delimiters"
Re: Mathed delimiters bug
On Fri, Apr 12, 2002 at 04:18:52PM +0200, Jean-Marc Lasgouttes wrote: > When trying to use M-m [ or M-m (, I get respectively > formulabase::LFUN_MATH_DELIM, arg: '[ ]' > can't parse delimeters from '[ ]' I see the first line, but not the second. And it inserts () and []. I just removed the spurious message. Andre' -- Those who desire to give up Freedom in order to gain Security, will not have, nor do they deserve, either one. (T. Jefferson)
