Calibre footnote viewer missing out words

[bookman156]

I noticed a curious thing inspecting footnotes in the footnotes panel. Some words are missed out, usually book titles done with a span class italic. The missing words are still missing when the footnote panel is undocked. But when the footnote is viewed in the main view the missing words are back and everything is fine. The EPUB validates and there is nothing wrong with the code at these points. The problem is also sporadic, not consistent. Similar text might be displayed fine.

Is this a known issue? I wish I had a Kindle Paperwhite to see whether it also happened there.

Edit: This is in v 2.52.

[kovidgoyal]

You need to attach a sample book showing the issue.

[bookman156]

Unfortunately, I can't as it's copyrighted. Would a screenshot be of use?

[AnemicOak]

Originally Posted by bookman156

I wish I had a Kindle Paperwhite to see whether it also happened there.

There's always Kindle Previewer.

[BetterRed]

Originally Posted by bookman156

Unfortunately, I can't as it's copyrighted. Would a screenshot be of use?

@bookman156

You could scramble it with Scramble Ebook and post it here

Or you could post here as bug and mark the post private Bugs : calibre

BR

[davidfor]

Kovid: This sounds like something that was discussed earlier in the year. Some strange things happen if the id the footnote refers to is not at the top level of tags. If the id is in a nested tag, then the text in tags that are further nested is lost. The thread this was discussed in is https://www.mobileread.com/forums/sho...d.php?t=270041.

[bookman156]

I've scrambled the EPUB and deleted unscrambled text manually (Chinese).

Check for example Chapter III (DEU in Table of contents)

notes 9 and 18

at the end of 18 in the footnote pane:

(Bsud, , Jyp. SE, x 219).

But same note in main view:

(Bsud, vc. kqv, Jyp. SE, x 219).

In note 9:

Footnote pane:

(Ywlfh, ) Zdeyh Ohkvrt opgfhgoeci ttp rmjc bwkg: ‘Ple ahiqfsn kz gzo qt uiu scbvotilz, gyhi vke xdfck cvxyx uzvvbqgb?’ (Pcosod, , h 616.)

but in main view as it should be:

(Ywlfh, Zxmt Imt: T Ihsowpp Neor oc Dpqzmke.) Zdeyh Ohkvrt opgfhgoeci ttp rmjc bwkg: ‘Ple ahiqfsn kz gzo qt uiu scbvotilz, gyhi vke xdfck cvxyx uzvvbqgb?’ (Pcosod, Nwb Lkpsv ym nta Ovsxf, h 616.)

For some reason it seems to happen to italic text.

There are quite a few notes where such things occur.

Sometimes the note number is missing in the footnote pane, but there in the main view, as in note 23 of same chapter.

[kovidgoyal]

@davidfor: The way it works is the tag on which the id is defined and all its children are included, upto the next page break. Also all its parents, but not its siblings. If you see behavior different than that I'd appreciate a sample showing that behavior.

[kovidgoyal]

@bookman156: Your scrambled book does not match what you are posting. There is nothing named DEU in the table fo contents and footnote number 9 is

9 Viaoigsnnvm, ‘Pfy Gphnwrhvblx ak xvd Ffdvfi,’ o 2.

which appears as exactly that, minus the number on the popup.

[bookman156]

I meant in the print version of the table of contents. In the other TOC Chapter III is SEG.

The actual chapter is: Eaxeaws ZGD

[kovidgoyal]

Again, I dont see what you are saying, footnote 9 in that chapter is

Quni, rt. osr, Vnz. R, ne 209–012. Oix Tythu ajpffcq bgmb bevsn lh xcvqgdfcup u agwx me Uvfaiyd 04 cg dde Nivxvmjtk: ‘Jyicjz yrx erbck hl hhmi cn gy dfeetkvnyjtka.’ Rsxphy igq drrmq oeqkl mqd fgev gk ‘rrymrvf qsv azbcy’, cdrhm zm zcnbzjshiq iowz fehjihfy 86/5 by lno Szile Nivwlcim, Emvl QU, Eoywlrd J, Lqimu 96. Xz slr Klsurnu-Yrztld (7ku ajlvexa) twm w 71 xfk q 326. Fqrtbfg gtlq knn ‘roipldlxyrtib rsvfrxuq pr pnnlh asmvcz’ gq stcgrjbj 06/6 (lm 450–476); ‘owz rlkdqwvl dlervshbe jd pcrtw pgi lcfwfjoxwtzlm vqkvmg’ wg 48/0 (c 871); ‘tywfvxqros rrx znsth th lcfzfh ciysem’ ij 9/5 (q 73). Frdmnlr 97 rx zvm twblt hamu lpympdp chyry. Lwejfvv Idplk jbd olrjcizii ncfy aen xyebe js jdm Jfczonou qb Zw ufq yaw qzilf xezn-zygsny nyrufmdb .04 qp oqj , v zjauem hwcdfvg bhof vl ioglcre, hhidn qg fcvhrxdolv vm: ‘Jyz ibdemnjgg yi mdq fgl, ifs iti lq uvycggpxtk?’ (Ywlfh, ) Zdeyh Ohkvrt opgfhgoeci ttp rmjc bwkg: ‘Ple ahiqfsn kz gzo qt uiu scbvotilz, gyhi vke xdfck cvxyx uzvvbqgb?’ (Pcosod, , h 616.)

[davidfor]

Originally Posted by kovidgoyal

@davidfor: The way it works is the tag on which the id is defined and all its children are included, upto the next page break. Also all its parents, but not its siblings. If you see behavior different than that I'd appreciate a sample showing that behavior.

Attached is the example I built for the other thread. The first chapter "Endnote 1" has footnote links to the same three footnotes. The footnotes are in the chapter "The Footnotes".

Clicking footnote 1 one shows:

[1] Note 1.

Note 2.Note 2a.Note 2b.

Clicking footnote 3 shows:

Note 2.[2]Note 2a.Note 2b. Note 2c.

Note 3.

The second line of the first footnote is the first line of the second. The "[2]" is missing from the first footnote. Or maybe the rest of the text is the problem.

[bookman156]

Originally Posted by kovidgoyal

Again, I dont see what you are saying, footnote 9 in that chapter is

Quni, rt. osr, Vnz. R, ne 209–012. Oix Tythu ajpffcq bgmb bevsn lh xcvqgdfcup u agwx me Uvfaiyd 04 cg dde Nivxvmjtk: ‘Jyicjz yrx erbck hl hhmi cn gy dfeetkvnyjtka.’ Rsxphy igq drrmq oeqkl mqd fgev gk ‘rrymrvf qsv azbcy’, cdrhm zm zcnbzjshiq iowz fehjihfy 86/5 by lno Szile Nivwlcim, Emvl QU, Eoywlrd J, Lqimu 96. Xz slr Klsurnu-Yrztld (7ku ajlvexa) twm w 71 xfk q 326. Fqrtbfg gtlq knn ‘roipldlxyrtib rsvfrxuq pr pnnlh asmvcz’ gq stcgrjbj 06/6 (lm 450–476); ‘owz rlkdqwvl dlervshbe jd pcrtw pgi lcfwfjoxwtzlm vqkvmg’ wg 48/0 (c 871); ‘tywfvxqros rrx znsth th lcfzfh ciysem’ ij 9/5 (q 73). Frdmnlr 97 rx zvm twblt hamu lpympdp chyry. Lwejfvv Idplk jbd olrjcizii ncfy aen xyebe js jdm Jfczonou qb Zw ufq yaw qzilf xezn-zygsny nyrufmdb .04 qp oqj , v zjauem hwcdfvg bhof vl ioglcre, hhidn qg fcvhrxdolv vm: ‘Jyz ibdemnjgg yi mdq fgl, ifs iti lq uvycggpxtk?’ (Ywlfh, ) Zdeyh Ohkvrt opgfhgoeci ttp rmjc bwkg: ‘Ple ahiqfsn kz gzo qt uiu scbvotilz, gyhi vke xdfck cvxyx uzvvbqgb?’ (Pcosod, , h 616.)

You have to compare this footnote you quote from the footnote pane, which has the text missing, with the same footnote viewed in the main view. Then you will see what I mean. The text that is missing is from the bottom few lines:

(Ywlfh, ) Zdeyh Ohkvrt opgfhgoeci ttp rmjc bwkg: ‘Ple ahiqfsn kz gzo qt uiu scbvotilz, gyhi vke xdfck cvxyx uzvvbqgb?’ (Pcosod, , h 616.)

After Ywlfh, and after Pcosod,. There is italic text there that is absent here, as you will see in the actual footnote as seen in the main view.

[kovidgoyal]

@bookman156: That happens because you have anchors *inside* the footnote that are pointed to by links in the main text. For example in footnote 9 you have <a id='page170'> Remove that and you will be fine.

The heuristic used to detect footnotes stops on anchors that are pointed to by other links in the text, on the theory that they could be the start of the next footnote. That said, I have changed the heuristic to only regard the anchor as a breaking point if it is not styled as an inline element.

@davidfor: I committed a fix for your issue, you should be fine now.

[davidfor]

Originally Posted by kovidgoyal

@davidfor: I committed a fix for your issue, you should be fine now.

Yes, that fixed it.