Re: LyX indexing anomaly SOLVED
Steve Litt wrote: The correct solution would be to put the | character in an ERT. That's not a very good solution because it forces the user to do extra work in every index range of every document, and also gravely increases the troubleshooting effort if the user messes up the nesting of |( and |). As I've explained on trac, ERT is the correct solution because | is an active character and thus native LaTeX (that is actually not yet natively supported by LyX). It is pure coincidence that it also works without ERT in T1. So T1 and non-ERT is the exception, ERT is the rule (unless we have native support). From my perspective as a book author, T1 would be preferable to ERT, and even my silly kludge script would be preferable. This only works if T1 is the desired font encoding. For any other than Latin script, it won't. So it's a pretty limited workaround. T1, ERT and my script are all workarounds. The SOLUTION would involve recognition by LyX that |( or |) at the end of a string inside an IDX tag defines a range, so do not convert the pipe symbol to \textbar etc. No, automatic fiddling behind the user's back is not an option. Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. Yes, something like this. Jürgen
Re: LyX indexing anomaly SOLVED
On Monday 31 August 2009 03:17:53 Jürgen Spitzmüller wrote: Steve Litt wrote: Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. Yes, something like this. Jürgen OK, then let's do it with something like this. SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt
Re: LyX indexing anomaly SOLVED
Steve Litt wrote: The correct solution would be to put the | character in an ERT. That's not a very good solution because it forces the user to do extra work in every index range of every document, and also gravely increases the troubleshooting effort if the user messes up the nesting of |( and |). As I've explained on trac, ERT is the correct solution because | is an active character and thus native LaTeX (that is actually not yet natively supported by LyX). It is pure coincidence that it also works without ERT in T1. So T1 and non-ERT is the exception, ERT is the rule (unless we have native support). From my perspective as a book author, T1 would be preferable to ERT, and even my silly kludge script would be preferable. This only works if T1 is the desired font encoding. For any other than Latin script, it won't. So it's a pretty limited workaround. T1, ERT and my script are all workarounds. The SOLUTION would involve recognition by LyX that |( or |) at the end of a string inside an IDX tag defines a range, so do not convert the pipe symbol to \textbar etc. No, automatic fiddling behind the user's back is not an option. Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. Yes, something like this. Jürgen
Re: LyX indexing anomaly SOLVED
On Monday 31 August 2009 03:17:53 Jürgen Spitzmüller wrote: Steve Litt wrote: Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. Yes, something like this. Jürgen OK, then let's do it with something like this. SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt
Re: LyX indexing anomaly SOLVED
Steve Litt wrote: So, if anyone has indexing problems and sees range type index code that looks like this: \index{chapter 2|(@chapter 2\textbar{}(} instead of this: \index{chapter 2|(} then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. The correct solution would be to put the | character in an ERT. THE ONLY REMAINING TASK is to understand the meaning of T1. What is it? Does this help? http://www.tex.ac.uk/cgi-bin/texfaq2html?label=whatenc It's the set of characters that you will be using in the document -- the set of characters you will be restricted to. For instance, the following is the T1 encoding: http://www.micropress-inc.com/fonts/encoding/t1.htm The following is OT1, which looks a heck of a lot like ASCII to me: http://www.micropress-inc.com/fonts/encoding/ot1.htm Here's where the plot thickens. I've read that if you select a specific font encoding like T1, it can change the font version that your document uses: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=fuzzy-T1 Could someone please explain the info in the preceding URL to me? In fact, a lot of my stuff IS fuzzy when viewed in xpdf. It happens that, for the T1 font encoding, some (in the past: all) LaTeX distributions use a bitmap default font. This bitmap font looks fuzzy in the PDF viewer (but not in the printout). The solution is to switch to a PostScript font such as Latin Modern. See this FAQ: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=type1T1 Jürgen
Re: LyX indexing anomaly SOLVED
On Sunday 30 August 2009 05:48:58 Jürgen Spitzmüller wrote: Steve Litt wrote: then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. The correct solution would be to put the | character in an ERT. That's not a very good solution because it forces the user to do extra work in every index range of every document, and also gravely increases the troubleshooting effort if the user messes up the nesting of |( and |). From my perspective as a book author, T1 would be preferable to ERT, and even my silly kludge script would be preferable. T1, ERT and my script are all workarounds. The SOLUTION would involve recognition by LyX that |( or |) at the end of a string inside an IDX tag defines a range, so do not convert the pipe symbol to \textbar etc. Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt
Re: LyX indexing anomaly SOLVED
Steve Litt wrote: So, if anyone has indexing problems and sees range type index code that looks like this: \index{chapter 2|(@chapter 2\textbar{}(} instead of this: \index{chapter 2|(} then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. The correct solution would be to put the | character in an ERT. THE ONLY REMAINING TASK is to understand the meaning of T1. What is it? Does this help? http://www.tex.ac.uk/cgi-bin/texfaq2html?label=whatenc It's the set of characters that you will be using in the document -- the set of characters you will be restricted to. For instance, the following is the T1 encoding: http://www.micropress-inc.com/fonts/encoding/t1.htm The following is OT1, which looks a heck of a lot like ASCII to me: http://www.micropress-inc.com/fonts/encoding/ot1.htm Here's where the plot thickens. I've read that if you select a specific font encoding like T1, it can change the font version that your document uses: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=fuzzy-T1 Could someone please explain the info in the preceding URL to me? In fact, a lot of my stuff IS fuzzy when viewed in xpdf. It happens that, for the T1 font encoding, some (in the past: all) LaTeX distributions use a bitmap default font. This bitmap font looks fuzzy in the PDF viewer (but not in the printout). The solution is to switch to a PostScript font such as Latin Modern. See this FAQ: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=type1T1 Jürgen
Re: LyX indexing anomaly SOLVED
On Sunday 30 August 2009 05:48:58 Jürgen Spitzmüller wrote: Steve Litt wrote: then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. The correct solution would be to put the | character in an ERT. That's not a very good solution because it forces the user to do extra work in every index range of every document, and also gravely increases the troubleshooting effort if the user messes up the nesting of |( and |). From my perspective as a book author, T1 would be preferable to ERT, and even my silly kludge script would be preferable. T1, ERT and my script are all workarounds. The SOLUTION would involve recognition by LyX that |( or |) at the end of a string inside an IDX tag defines a range, so do not convert the pipe symbol to \textbar etc. Or, if that's impossible, when the LyX user requests an index tag, pop up a dialog box asking the name and whether it's begin-range, end-range, or this-location. SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt
Re: LyX indexing anomaly SOLVED
Thanks Vincent. Please continue reading at the bottom... On Thursday 27 August 2009 04:00:43 you wrote: Hi all, My copy of LyX 1.6.3 errors out on indexing because when LyX exports to LaTeX, it outputs normal range type indexing entries (the ones ending in a pipe symbol and a parenthese) with something like this: \index{mysection|(@mysection\textbar(} It should have looked like this: \index{mysection|(} Steve, In case you didn't notice this mail (devel-list), I will repeat it here. What's your Tools-Preferences-Output-LaTeX-Tex Encoding option ? If this doesn't read T1, put it in there and I think this problem will go away At least, I can reproduce it also when setting the document language to greek. Then this 'trick' with '|)' doesn't work any longer. That's a bug too. Vincent Tools-Preferences-Output-LaTeX-Tex Encoding was set to default on my system. After I set it to T1, the extraneous text no longer occurred. I set it back to default, and the extraneous text came back. So, if anyone has indexing problems and sees range type index code that looks like this: \index{chapter 2|(@chapter 2\textbar{}(} instead of this: \index{chapter 2|(} then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. THE ONLY REMAINING TASK is to understand the meaning of T1. What is it? It's the set of characters that you will be using in the document -- the set of characters you will be restricted to. For instance, the following is the T1 encoding: http://www.micropress-inc.com/fonts/encoding/t1.htm The following is OT1, which looks a heck of a lot like ASCII to me: http://www.micropress-inc.com/fonts/encoding/ot1.htm Here's where the plot thickens. I've read that if you select a specific font encoding like T1, it can change the font version that your document uses: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=fuzzy-T1 Could someone please explain the info in the preceding URL to me? In fact, a lot of my stuff IS fuzzy when viewed in xpdf. Vincent and everyone else -- thanks so much for your help! SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt
Re: LyX indexing anomaly SOLVED
Thanks Vincent. Please continue reading at the bottom... On Thursday 27 August 2009 04:00:43 you wrote: Hi all, My copy of LyX 1.6.3 errors out on indexing because when LyX exports to LaTeX, it outputs normal range type indexing entries (the ones ending in a pipe symbol and a parenthese) with something like this: \index{mysection|(@mysection\textbar(} It should have looked like this: \index{mysection|(} Steve, In case you didn't notice this mail (devel-list), I will repeat it here. What's your Tools-Preferences-Output-LaTeX-Tex Encoding option ? If this doesn't read T1, put it in there and I think this problem will go away At least, I can reproduce it also when setting the document language to greek. Then this 'trick' with '|)' doesn't work any longer. That's a bug too. Vincent Tools-Preferences-Output-LaTeX-Tex Encoding was set to default on my system. After I set it to T1, the extraneous text no longer occurred. I set it back to default, and the extraneous text came back. So, if anyone has indexing problems and sees range type index code that looks like this: \index{chapter 2|(@chapter 2\textbar{}(} instead of this: \index{chapter 2|(} then the most likely solution is to set Tools-Preferences-Output-LaTeX-Tex to T1. THE ONLY REMAINING TASK is to understand the meaning of T1. What is it? It's the set of characters that you will be using in the document -- the set of characters you will be restricted to. For instance, the following is the T1 encoding: http://www.micropress-inc.com/fonts/encoding/t1.htm The following is OT1, which looks a heck of a lot like ASCII to me: http://www.micropress-inc.com/fonts/encoding/ot1.htm Here's where the plot thickens. I've read that if you select a specific font encoding like T1, it can change the font version that your document uses: http://www.tex.ac.uk/cgi-bin/texfaq2html?label=fuzzy-T1 Could someone please explain the info in the preceding URL to me? In fact, a lot of my stuff IS fuzzy when viewed in xpdf. Vincent and everyone else -- thanks so much for your help! SteveT Steve Litt Recession Relief Package http://www.recession-relief.US Twitter: http://www.twitter.com/stevelitt