Re: [NTG-context] [luatex] Problem with luaotfload and positioning of primes

newer
Alignment, indenting and font...

older
Problem with List of Abbreviations

Philipp Gesang

28 Jul 2014 28 Jul '14

10:56 p.m.

(CC’ing the Context list.) ·········

...

Do you have a prediction about how long it will take before a fix is released?

No, sorry. What I do have though is the culprit: c70b2e466bbb07e976f8684d56ae4e5398169e7d is the first bad commit commit c70b2e466bbb07e976f8684d56ae4e5398169e7d Author: Hans Hagen Date: Sat Mar 22 15:31:00 2014 +0100 beta 2014.03.22 15:31 ... See the commit diff [0]. @Hans: Here is a minimal example for Luatex-Plain: https://bitbucket.org/phg/lua-la-tex-tests/src/8d48db1435bbd70888381a6cacd22... With the last good commit (244a447.. dated 2014.02.14 in the Garden repo) both primes are typeset correctly. With c70b2e46.. -- which is unfortunately the first snapshot after Marious’ mirror went dark -- the second one doesn’t take the ssty feature into account. Best regards, Philipp [0] https://bitbucket.org/phg/context-mirror/commits/c70b2e466bbb07e976f8684d56a...

Attachments:

attachment.sig (application/pgp-signature — 819 bytes)

Show replies by date

Philipp Gesang

28 Jul 28 Jul

11:35 p.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

·········

...

(CC’ing the Context list.)

·········

...
Do you have a prediction about how long it will take before a fix is released?

No, sorry. What I do have though is the culprit:

c70b2e466bbb07e976f8684d56ae4e5398169e7d is the first bad commit

Seems to be related to the feature precedence code. If I revert the related changes as in the attached patch, the bug vanishes (and so does te feature ordering …).

Philipp Gesang

29 Jul 29 Jul

7:01 a.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

Good morning, Hans! Thanks for the quick fix! Best, Philipp ·········

...

·········

...
(CC’ing the Context list.)

·········

...
Do you have a prediction about how long it will take before a fix is released?

No, sorry. What I do have though is the culprit:

c70b2e466bbb07e976f8684d56ae4e5398169e7d is the first bad commit

Seems to be related to the feature precedence code. If I revert the related changes as in the attached patch, the bug vanishes (and so does te feature ordering …).

...

diff --git a/src/luaotfload-fontloader.lua b/src/luaotfload-fontloader.lua index 293a319..719a1f7 100644 --- a/src/luaotfload-fontloader.lua +++ b/src/luaotfload-fontloader.lua @@ -9152,34 +9152,26 @@ local function featuresinitializer(tfmdata,value) local language=properties.language local basesubstitutions=rawdata.resources.features.gsub local basepositionings=rawdata.resources.features.gpos - if basesubstitutions or basepositionings then - local sequences=tfmdata.resources.sequences - for s=1,#sequences do - local sequence=sequences[s] - local sfeatures=sequence.features - if sfeatures then - local order=sequence.order - if order then - for i=1,#order do - local feature=order[i] - if features[feature] then - local validlookups,lookuplist=collectlookups(rawdata,feature,script,language) - if not validlookups then - elseif basesubstitutions and basesubstitutions[feature] then - if trace_preparing then - report_prepare("filtering base feature %a for %a",feature,fullname) - end - applybasemethod("preparesubstitutions",tfmdata,feature,value,validlookups,lookuplist) - registerbasefeature(feature,value) - elseif basepositionings and basepositionings[feature] then - if trace_preparing then - report_prepare("filtering base feature %a for %a",feature,fullname) - end - applybasemethod("preparepositionings",tfmdata,feature,features[feature],validlookups,lookuplist) - registerbasefeature(feature,value) - end - end - end + if basesubstitutions then + for feature,data in next,basesubstitutions do + local value=features[feature] + if value then + local validlookups,lookuplist=collectlookups(rawdata,feature,script,language) + if validlookups then + applybasemethod("preparesubstitutions",tfmdata,feature,value,validlookups,lookuplist) + registerbasefeature(feature,value) + end + end + end + end + if basepositionings then + for feature,data in next,basepositionings do + local value=features[feature] + if value then + local validlookups,lookuplist=collectlookups(rawdata,feature,script,language) + if validlookups then + applybasemethod("preparepositionings",tfmdata,feature,features[feature],validlookups,lookuplist) + registerbasefeature(feature,value) end end end

...

___________________________________________________________________________________ If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context webpage : http://www.pragma-ade.nl / http://tex.aanhet.net archive : http://foundry.supelec.fr/projects/contextrev/ wiki : http://contextgarden.net ___________________________________________________________________________________

Hans Hagen

10:35 a.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

On 7/29/2014 7:01 AM, Philipp Gesang wrote:

...

Thanks for the quick fix!

You were just lucky that I saw it pop up as incoming mail message (which I somehow cannot turn off) because normally all mails directed a the luatex list is filtered away to its own bin as it seldom relates to context. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Philipp Gesang

7:18 p.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

·········

...

On 7/29/2014 7:01 AM, Philipp Gesang wrote:

...
Thanks for the quick fix!

You were just lucky that I saw it pop up as incoming mail message (which I somehow cannot turn off) because normally all mails directed a the luatex list is filtered away to its own bin as it seldom relates to context.

True. Though I put the Context list in the Cc as soon as I was certain the bug originates in the fontloader. Best, Philipp

Hans Hagen

7:20 p.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

On 7/29/2014 7:18 PM, Philipp Gesang wrote:

...

·········

...
On 7/29/2014 7:01 AM, Philipp Gesang wrote:

...
Thanks for the quick fix!

You were just lucky that I saw it pop up as incoming mail message (which I somehow cannot turn off) because normally all mails directed a the luatex list is filtered away to its own bin as it seldom relates to context.

True. Though I put the Context list in the Cc as soon as I was certain the bug originates in the fontloader.

swap to and cc then (the filter works on to -) Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | voip: 087 875 68 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------

Philipp Gesang

8:03 p.m.

New subject: [luatex] Problem with luaotfload and positioning of primes

·········

...

On 7/29/2014 7:18 PM, Philipp Gesang wrote:

...
·········

...
On 7/29/2014 7:01 AM, Philipp Gesang wrote:

...
Thanks for the quick fix!

You were just lucky that I saw it pop up as incoming mail message (which I somehow cannot turn off) because normally all mails directed a the luatex list is filtered away to its own bin as it seldom relates to context.

True. Though I put the Context list in the Cc as soon as I was certain the bug originates in the fontloader.

swap to and cc then (the filter works on to -)

Noted!

3732

Age (days ago)

3733

Last active (days ago)

List overview

Download

6 comments

3 participants

participants (3)

Hans Hagen
Philipp Gesang
Philipp Gesang