Use span instead of div in the text layer

[PalmerAL]

This improves copy/pasting text content since it reduces the amount of unnecessary newlines.

[timvandermeij]

/botio test

[pdfjsbot]

From: Bot.io (Windows)

---

Received

Command cmd_test from @timvandermeij received. Current queue size: 0

Live output at: http://54.215.176.217:8877/e2ac2c1d78b0882/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Received

Command cmd_test from @timvandermeij received. Current queue size: 0

Live output at: http://54.67.70.0:8877/3b9b3895a21e26f/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Failed

Full output at http://54.67.70.0:8877/3b9b3895a21e26f/output.txt

Total script time: 19.21 mins

• Font tests: Passed
• Unit tests: Passed
• Regression tests: FAILED

Image differences available at: http://54.67.70.0:8877/3b9b3895a21e26f/reftest-analyzer.html#web=eq.log

[pdfjsbot]

From: Bot.io (Windows)

---

Failed

Full output at http://54.215.176.217:8877/e2ac2c1d78b0882/output.txt

Total script time: 25.07 mins

• Font tests: Passed
• Unit tests: Passed
• Regression tests: FAILED

Image differences available at: http://54.215.176.217:8877/e2ac2c1d78b0882/reftest-analyzer.html#web=eq.log

[PalmerAL]

It looks like the CSS for the <span> elements isn't being applied for some reason, so they aren't positioned correctly. I tested a couple of the PDF's that are failing, and they all seem to work fine locally, so I can't think of any reason why they would be failing in the tests. Is there something I'm missing?

[timvandermeij]

Ah, I think you also need to change https://github.com/mozilla/pdf.js/blob/master/test/text_layer_test.css#L25 for the tests.

[PalmerAL]

Thanks for pointing that out! It looks like there's some other CSS rules that depend on the items being divs as well, I think I've fixed them all in 45279d2.

[timvandermeij]

/botio test

[pdfjsbot]

From: Bot.io (Windows)

---

Received

Command cmd_test from @timvandermeij received. Current queue size: 1

Live output at: http://54.215.176.217:8877/779528733d58856/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Received

Command cmd_test from @timvandermeij received. Current queue size: 0

Live output at: http://54.67.70.0:8877/219385726c29ce4/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Success

Full output at http://54.67.70.0:8877/219385726c29ce4/output.txt

Total script time: 19.11 mins

• Font tests: Passed
• Unit tests: Passed
• Regression tests: Passed

[pdfjsbot]

From: Bot.io (Windows)

---

Success

Full output at http://54.215.176.217:8877/779528733d58856/output.txt

Total script time: 24.56 mins

• Font tests: Passed
• Unit tests: Passed
• Regression tests: Passed

[timvandermeij]

/botio-linux preview

[pdfjsbot]

From: Bot.io (Linux m4)

---

Received

Command cmd_preview from @timvandermeij received. Current queue size: 0

Live output at: http://54.67.70.0:8877/7daa5ccb06f81eb/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Success

Full output at http://54.67.70.0:8877/7daa5ccb06f81eb/output.txt

Total script time: 2.84 mins

Published

• Viewer: http://54.67.70.0:8877/7daa5ccb06f81eb/web/viewer.html

[timvandermeij]

Thank you for your contribution; I think this is a good improvement! I copy/pasted the following paragraph of the Tracemonkey paper with and without your patch applied. Before applying your patch the result was:

A
typed trace
is a trace annotated with a type for every variable
(including temporaries) on the trace. A typed trace also has an entry
type map
giving the required types for variables used on the trace
before they are defined. For example, a trace could have a type map
(x: int, b: boolean)
, meaning that the trace may be entered
only if the value of the variable
x
is of type
int
and the value of
b
is of type
boolean
. The entry type map is much like the signature
of a function.

In contrast, after applying your patch the result was:

Atyped traceis a trace annotated with a type for every variable(including temporaries) on the trace. A typed trace also has an entrytype mapgiving the required types for variables used on the tracebefore they are defined. For example, a trace could have a type map(x: int, b: boolean), meaning that the trace may be enteredonly if the value of the variablexis of typeintand the value ofbis of typeboolean. The entry type map is much like the signatureof a function.

Obviously there is enough room for improvement, but at least the unnecessary newlines are now gone.

Unrelated note: this is also the first time I used GitHub's squash-and-commit feature and it worked really well since it also allows you to choose the new commit message for the squashed commits.

[PalmerAL]

@timvandermeij Thanks for merging this! Any thoughs on what to do with #10172? If you think the complexity of that isn't worth it (which I completely understand), one alternative I've been thinking about is whether it would be possible to unconditionally insert a space before every text layer item, which would fix things like your example above; if that sounds OK to you, I can spend some time investigating that further.

[timvandermeij]

@PalmerAL Yes, I think that one is still a bit too complex, at least in its current form. Perhaps we can do it in smaller steps, i.e., first try to get the spaces right and after that perhaps the newlines. If you can find a solution where the spaces are correct and no double ones are inserted, feel free to create a PR for that and we'll review. I'll close the current PR for you. Thank you for looking into this; if we take small steps here I'm sure we can make more improvements.