glftpos=right option broken for nlevel gloss

[peteraisher]

Side by side gloss and translation is broken when using glstyle=nlevel. The glftpos=right option is ignored if it comes before glstyle=nlevel and causes errors if it comes after.

It looks like \glft is redefined by the gfltpos option on lines expex.tex:1313 and 1317, but also by the glstyle option on lines 715 and 720, but in a way that is not reconcilable.

Sadly my understanding ends there!

[frampton]

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

[peteraisher]

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

[frampton]

Hi Peter,

Most people who ask questions about ExPex are linguists, so I googled you to find out where you were. Needless to say, I was a bit surprised. I'm not a musician, but I also pedaled to work on a bike, for similar reasons. And did long rides. I also started out in physics. So there is something in common.

How did you find out about ExPex and what is your particular interest in it? How do you plan to use it? I am always on the lookout for interesting applications. The next edition of the User's Manual will contain several extensive examples.

Not many linguists use the nlevel gloss style, because it is not the way they were brought up to lay out glosses. They tend to think of a gloss as a big array, rather than a sequence of words with things attached to them. Daniel Harbour (Queen Mary University in London) is one of the few. Since it is not used much, it is not surprising that all the bugs have not been found. So thanks for finding one. I'm right in the middle of racking my brains about something else, but I will get to fixing it sometime soon.

Best, John

On 1/4/2015 4:03 AM, peteraisher wrote:

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2#issuecomment-68626075.

[peteraisher]

Hi John,

Nice to know I'm easily discoverable!

My interest in ExPex stems from the practice which the more serious/scholarly singers I know undertake of getting to understand every word of any opera libretto or song text which they may be singing. There are a series of books by Castel for many of the major operas which provide a three-line "gloss" of original language, IPA for sung pronunciation, and English word by word.

Whilst these are useful, many less well-known operas are not available, and at any rate I find it informative to produce my own translations, especially as it's mostly poetry rather than prose.

Because I also normally pencil a word-by-word and the odd phonetic reminder into the score, it's fairly natural for me to use the n-level style, as, with musical dynamics, articulations, etc, I am sort of thinking of "words with bits attached" just like "notes with bits attached".

It also means I can paste a single language libretto into a text editor and use some find/replace magic for little, but constant words like

Peter Aisher | Tenor www.peteraisher.com

On 4 Jan 2015, at 13:04, John Frampton notifications@github.com wrote:

Hi Peter,

Most people who ask questions about ExPex are linguists, so I googled you to find out where you were. Needless to say, I was a bit surprised. I'm not a musician, but I also pedaled to work on a bike, for similar reasons. And did long rides. I also started out in physics. So there is something in common.

How did you find out about ExPex and what is your particular interest in it? How do you plan to use it? I am always on the lookout for interesting applications. The next edition of the User's Manual will contain several extensive examples.

Not many linguists use the nlevel gloss style, because it is not the way they were brought up to lay out glosses. They tend to think of a gloss as a big array, rather than a sequence of words with things attached to them. Daniel Harbour (Queen Mary University in London) is one of the few. Since it is not used much, it is not surprising that all the bugs have not been found. So thanks for finding one. I'm right in the middle of racking my brains about something else, but I will get to fixing it sometime soon.

Best, John

On 1/4/2015 4:03 AM, peteraisher wrote:

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2#issuecomment-68626075.

— Reply to this email directly or view it on GitHub.

[peteraisher]

Hi John,

Nice to know I'm easily discoverable!

My interest in ExPex stems from the practice which the more serious/scholarly singers I know undertake of getting to understand every word of any opera libretto or song text which they may be singing. There are a series of books by Castel for many of the major operas which provide a three-line "gloss" of original language, IPA for sung pronunciation, and English word by word.

Whilst these are useful, many less well-known operas are not available, and at any rate I find it informative to produce my own translations, especially as it's mostly poetry rather than prose.

Because I also normally pencil a word-by-word and the odd phonetic reminder into the score, it's fairly natural for me to use the n-level style, as, with musical dynamics, articulations, etc, I am sort of thinking of "words with bits attached" just like "notes with bits attached".

It also means I can paste a single language libretto into a text editor and use some find/replace magic for little, but constant words like "the" and "and" etc.

Thanks for looking into it and I'm glad my unusualness has been helpful in spotting a bug!

All best,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 4 Jan 2015, at 13:04, John Frampton notifications@github.com wrote:

Hi Peter,

Most people who ask questions about ExPex are linguists, so I googled you to find out where you were. Needless to say, I was a bit surprised. I'm not a musician, but I also pedaled to work on a bike, for similar reasons. And did long rides. I also started out in physics. So there is something in common.

How did you find out about ExPex and what is your particular interest in it? How do you plan to use it? I am always on the lookout for interesting applications. The next edition of the User's Manual will contain several extensive examples.

Not many linguists use the nlevel gloss style, because it is not the way they were brought up to lay out glosses. They tend to think of a gloss as a big array, rather than a sequence of words with things attached to them. Daniel Harbour (Queen Mary University in London) is one of the few. Since it is not used much, it is not surprising that all the bugs have not been found. So thanks for finding one. I'm right in the middle of racking my brains about something else, but I will get to fixing it sometime soon.

Best, John

On 1/4/2015 4:03 AM, peteraisher wrote:

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2#issuecomment-68626075.

— Reply to this email directly or view it on GitHub.

[frampton]

Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2) To: frampton/ExPex reply@reply.github.com Cc: Bcc: Reply-To: -=-=-=-=-=-=-=-=-=# Don't remove this line #=-=-=-=-=-=-=-=-=- Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2) To: frampton/ExPex reply@reply.github.com Cc: Bcc: Reply-To: -=-=-=-=-=-=-=-=-=# Don't remove this line #=-=-=-=-=-=-=-=-=-

I have discovered what the problem is, but have not fixed it yet. It seems that I added the glftpos options before I added the nlevel style glosses. So setting glftpos assumes that the gloss is in the older wrap style.

I don't think it will be very hard to fix. That's next.

After I get it working, perhaps you can send me a sample glossed libretto.

Best, John

If you want, you can send email directly to: j.frampton@neu.edu

[frampton]

Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2) To: frampton/ExPex reply@reply.github.com Cc: Bcc: Reply-To: -=-=-=-=-=-=-=-=-=# Don't remove this line #=-=-=-=-=-=-=-=-=-

glftpos=right is now compatible with glstyle=nlevel

The new version is on Github, but I am also attaching the new files. Of course, I am anxiously awaiting reports of new bugs that killing some of the old ones uncovers.

Enjoy, John \ProvidesPackage{expex}% [2015/01/06 v5.0d LaTex wrapper for expex.tex (JF)] \RequirePackage{xkeyval} \input expex \endinput

[frampton]

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075.

[peteraisher]

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

[frampton]

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752.

[peteraisher]

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

[frampton]

A shock! I certainly see the problem.

I’ll fix it tonight or tomorrow.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Saturday, January 10, 2015 6:32 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69452744.

[frampton]

Hi Peter,

I think the attached will fix the problem. It hasn’t been pushed to github yet. The glossing code is beginning to look like spaghetti. I have to go in and clean up the code before I push it to github and put it on CTAN.

As usual, I await your entomological comments.

Best, John

From: peteraisher [mailto:notifications@github.com] Sent: Saturday, January 10, 2015 6:32 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69452744.

[frampton]

Any further comments before I give it my blessings and send it off to CTAN?

From: peteraisher [mailto:notifications@github.com] Sent: Saturday, January 10, 2015 6:32 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69452744.

[peteraisher]

Hi, sorry to take so long to get back to you, I've been rather snowed under. Don't let me hold you back from sending it to CTAN, I'm sure your testing is more thorough than mine!

Peter Aisher | Tenor www.peteraisher.com

On 22 Jan 2015, at 22:03, John Frampton notifications@github.com wrote:

Any further comments before I give it my blessings and send it off to CTAN?

From: peteraisher [mailto:notifications@github.com] Sent: Saturday, January 10, 2015 6:32 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69452744. — Reply to this email directly or view it on GitHub.

[frampton]

OK, thanks.

Maybe sometime you could send me an annotated libretto, it would be a good example to put in a future User’s Guide. It does not have to be in Tex, I can do that and send it back to you to see if I got it right.

From: peteraisher [mailto:notifications@github.com] Sent: Tuesday, January 27, 2015 8:41 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi, sorry to take so long to get back to you, I've been rather snowed under. Don't let me hold you back from sending it to CTAN, I'm sure your testing is more thorough than mine!

Peter Aisher | Tenor www.peteraisher.com

On 22 Jan 2015, at 22:03, John Frampton notifications@github.com wrote:

Any further comments before I give it my blessings and send it off to CTAN?

From: peteraisher [mailto:notifications@github.com] Sent: Saturday, January 10, 2015 6:32 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Sorry, I should have mentioned that I'm using \exdisplay rather than \ex as I don't need/want the numbers and indentation. Your example (and mine if I switch to \ex) works perfectly. I had assumed that the environment was internally the same (i.e. \ex prints an example number and then typesets the remainder exactly as \exdisplay) which is why I didn't mention it.

For a minimal example, simply change yours to \exdisplay.

Sorry if I've led you down a blind alley!

Thanks,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 9 Jan 2015, at 12:59, John Frampton notifications@github.com wrote:

Hi Peter,

I’m not sure what the problem is. I’m attaching a demo file that seems to do what is expected. Perhaps you could run it and tell me if there is a problem for you. It uses no packages other than expex. If there is no problem with the demo file, you’ll have to send me a minimal example which shows the problem. Minimal means you should make it as simple as possible and use no packages other than those that are needed to illustrate the problem. I am assuming you use Latex, so I sent a Latex file.

• John

From: peteraisher [mailto:notifications@github.com] Sent: Thursday, January 08, 2015 5:50 PM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Hi John,

I had a play around but there seems to be a par break getting in after the gloss itself when using glftpos=right — but the next paragraph containing the free translation is indented by the amount of sssep if that helps with diagnosis?

In the meantime, I've had success by uncommenting the bit marked experimental in ExPex.tex and enclosing the contents of glft in {}, which works for me with no issues at the moment.

Wasn't sure if it was some other package interfering, and didn't want to bother you until I'd tried a minimal example, which I now have, and the par break remains.

From the code it looks like it's happening when the code for nlevel+right switches over to nlevel+below at the end of the gloss, which makes sense as there's a \par inside the definition of \gl@nlevel@below@end, which is called from inside \gl@nlevel@right@medial — could that be the case?

All best, and thanks again for your help,

Peter

Peter Aisher | Tenor www.peteraisher.com

On 8 Jan 2015, at 20:48, John Frampton notifications@github.com wrote:

Did you get a chance to look at the ExPex fix?

From: peteraisher [mailto:notifications@github.com] Sent: Sunday, January 04, 2015 4:04 AM To: frampton/ExPex Cc: Frampton, John Subject: Re: [ExPex] glftpos=right option broken for nlevel gloss (#2)

Thanks for the quick response!

Peter Aisher | Tenor www.peteraisher.com

On 3 Jan 2015, at 19:58, John Frampton notifications@github.com wrote:

I will look into it and see what is going on.

On 1/3/2015 9:16 AM, peteraisher wrote:

Side by side gloss and translation is broken when using |glstyle=nlevel|. The |glftpos=right| option is ignored if it comes before |glstyle=nlevel| and causes errors if it comes after.

It looks like |\glft| is redefined by the |gfltpos| option on lines |expex.tex:1313| and |1317|, but also by the |glstyle| option on lines |715| and |720|, but in a way that is not reconcilable.

Sadly my understanding ends there!

— Reply to this email directly or view it on GitHub https://github.com/frampton/ExPex/issues/2.

— Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-68626075. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69264752. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-69452744. — Reply to this email directly or view it on GitHub.

— Reply to this email directly or view it on GitHubhttps://github.com/frampton/ExPex/issues/2#issuecomment-71648849.