XFA - Fix auto-sized fields (bug 1722030) #13806

calixteman · 2021-07-27T16:47:50Z

In order to better compute text fields size, use line height with no gaps (and consequently guessed height for text are slightly better in general).
Fix default background color in fields.

calixteman · 2021-07-27T16:49:02Z

A lot of failures are expected, overall it's an improvement, but of course if something is really bad, tell me.

/botio test

pdfjsbot · 2021-07-27T16:49:03Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/393c30e0644ab34/output.txt

pdfjsbot · 2021-07-27T16:49:03Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/5cb8e8165bab76b/output.txt

src/core/xfa/template.js

pdfjsbot · 2021-07-27T17:15:45Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/393c30e0644ab34/output.txt

Total script time: 26.70 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 619
  different ref/snapshot: 308
  different first/second rendering: 1

Image differences available at: http://54.67.70.0:8877/393c30e0644ab34/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-27T17:22:31Z

There are some timeouts.

calixteman · 2021-07-27T17:25:53Z

/botio test

pdfjsbot · 2021-07-27T17:25:54Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/6b6153ea8532336/output.txt

pdfjsbot · 2021-07-27T17:25:54Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/5bd6131c2b95a4e/output.txt

pdfjsbot · 2021-07-27T17:27:13Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/5cb8e8165bab76b/output.txt

Total script time: 38.16 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 6
  different ref/snapshot: 412
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/5cb8e8165bab76b/reftest-analyzer.html#web=eq.log

pdfjsbot · 2021-07-27T17:59:12Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/6b6153ea8532336/output.txt

Total script time: 33.29 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 6
  different ref/snapshot: 419

Image differences available at: http://54.67.70.0:8877/6b6153ea8532336/reftest-analyzer.html#web=eq.log

pdfjsbot · 2021-07-27T17:59:34Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/5bd6131c2b95a4e/output.txt

Total script time: 32.34 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 601
  different ref/snapshot: 366
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/5bd6131c2b95a4e/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-27T18:02:24Z

Still some errors

calixteman · 2021-07-27T18:04:44Z

/botio test

pdfjsbot · 2021-07-27T18:04:46Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/042379177ea3723/output.txt

pdfjsbot · 2021-07-27T18:04:46Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/fcd0644bf70e6fc/output.txt

pdfjsbot · 2021-07-27T18:36:47Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/fcd0644bf70e6fc/output.txt

Total script time: 32.01 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 609
  different ref/snapshot: 324
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/fcd0644bf70e6fc/reftest-analyzer.html#web=eq.log

pdfjsbot · 2021-07-27T18:38:15Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/042379177ea3723/output.txt

Total script time: 33.48 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  different ref/snapshot: 421
  different first/second rendering: 2

Image differences available at: http://54.67.70.0:8877/042379177ea3723/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-27T20:30:22Z

There are some errors on Windows, maybe intermittent?

On firefox-xfa_bug1718670_1-page3, there is a field which was highlighted and isn't anymore, is it still fillable?
Same for firefox-xfa_imm5257e-page1, firefox-xfa_bug1716980-page1 and following pages of the same PDF, firefox-xfa_annual_expense_report-page5, firefox-xfa_issue13556-page3, firefox-xfa_issue13213-page5, firefox-xfa_issue13213-page6, firefox-xfa_imm1295e-page1, firefox-xfa_issue13500-page1, firefox-xfa_issue13500-page2.

A button at the end of firefox-issue13751-page1 disappeared.

Some text in firefox-xfa_issue13500-page1 is now hidden behind a field.

- In order to better compute text fields size, use line height with no gaps (and consequently guessed height for text are slightly better in general). - Fix default background color in fields.

calixteman · 2021-07-28T08:22:04Z

About the field colors:

some fields have no <border><fill... which means that by default they're transparent and in this case the bg color is taken from the css (unfocused: violet & focused: transparent);
some fields have <border><fill/>... which means that the background color is white by default and then the bg color is set through the style attribute. Then colors from the css are overwritten and there are no more distinction between focused/unfocused.

There are several solutions here:

do nothing: it's the actual state
add some js callbacks to change the colors on focus/blur events;
add a background-image: 1x1 with the violet-transparent color and remove it when the field has the focus;
remove this violet background which is out of specs but it help to see fields;
other ideas ??

About firefox-issue13751-page1: the button moved on the next page but the test is only with first page (see lastPage attribute in test_manifest.json). Looking at this pdf, it has some layout issues but unrelated to this patch.
About firefox-xfa_issue13500-page1: it's on purpose: I checked in Acrobat & Masterpdf: some checkboxes are behind the text field.

Snuffleupagus · 2021-07-28T08:31:06Z

There are several solutions here:

do nothing: it's the actual state

That'd be my definite preference, since it seems like the simplest approach; unless this is a widespread problem in practice.

add some js callbacks to change the colors on focus/blur events;

add a background-image: 1x1 with the violet-transparent color and remove it when the field has the focus;

All of these ideas sound like they'd easily add a bunch of complexity (and overhead), thus negatively impacting code readability/maintainability.

marco-c · 2021-07-28T08:34:43Z

That'd be my definite preference, since it seems like the simplest approach; unless this is a widespread problem in practice.

My worry is that if some fields have it and some fields don't, users might think the white ones are not fillable. It seems to be occurring in many PDFs (Canadian ones included, and they're between the most important we want to support).

calixteman · 2021-07-28T08:53:12Z

My opinion as a user is that we must be consistent.
So either we put this violet bg on every fillable field in trying to minimize the cost (memory, perf) or we remove it from everywhere.
But having a mix is a bit confusing: it gives the impress that fields haven't the same role, behaviour, whatever...
We already had this discussion: #13006 (which didn't land).
About the background-image idea: it should pretty simple (I didn't try it for now):

.xfaTextfield {
   background-image: url('a.png'); /* a.png is a 1x1 image with a transparent blue */
}
.xfaTextfield:focus {
   background-image: none;
}

I have no idea of the impact on memory/perf but from a code pov I'd say it isn't that painful to maintain

calixteman · 2021-07-28T09:37:34Z

/botio test

pdfjsbot · 2021-07-28T09:37:35Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/7f4fb57d1503e0c/output.txt

pdfjsbot · 2021-07-28T09:37:35Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/4d53839184c0520/output.txt

pdfjsbot · 2021-07-28T10:09:35Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/7f4fb57d1503e0c/output.txt

Total script time: 31.99 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 610
  different ref/snapshot: 403
  different first/second rendering: 2

Image differences available at: http://3.101.106.178:8877/7f4fb57d1503e0c/reftest-analyzer.html#web=eq.log

pdfjsbot · 2021-07-28T10:11:05Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/4d53839184c0520/output.txt

Total script time: 33.49 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  different ref/snapshot: 539
  different first/second rendering: 2

Image differences available at: http://54.67.70.0:8877/4d53839184c0520/reftest-analyzer.html#web=eq.log

calixteman · 2021-07-28T11:58:58Z

I replaced png image by an inlined svg one: this way we can easily modify it if we need:
e64078c#diff-3047fd328b0593becf1d475dceb17eed893613549d26c537cb9d105a113106afR17

Thanks to that change, the reftest are fine with one exception (at least locally): the one for xfa_filled_imm1344e has a bg color issue but only for the first page and only with Firefox.
Maybe it's because it's the first test ran...

calixteman · 2021-07-28T11:59:07Z

/botio test

pdfjsbot · 2021-07-28T11:59:08Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/092c437812770d9/output.txt

pdfjsbot · 2021-07-28T11:59:08Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/9489a0012503d14/output.txt

pdfjsbot · 2021-07-28T12:32:35Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/092c437812770d9/output.txt

Total script time: 33.43 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 4
  different ref/snapshot: 468
  different first/second rendering: 2

Image differences available at: http://54.67.70.0:8877/092c437812770d9/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-28T12:37:14Z

Thanks to that change, the reftest are fine with one exception (at least locally): the one for xfa_filled_imm1344e has a bg color issue but only for the first page and only with Firefox.
Maybe it's because it's the first test ran...

It is happening with firefox-xfa_imm1295e-page1 too, is it only a reftest issue or is it occurring in the actual viewer too?

pdfjsbot · 2021-07-28T12:37:33Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/9489a0012503d14/output.txt

Total script time: 38.40 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  errors: 4
  different ref/snapshot: 461
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/9489a0012503d14/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-28T12:38:15Z

There are also a few errors.

…tency

calixteman · 2021-07-28T12:53:25Z

Thanks to that change, the reftest are fine with one exception (at least locally): the one for xfa_filled_imm1344e has a bg color issue but only for the first page and only with Firefox.
Maybe it's because it's the first test ran...

It is happening with firefox-xfa_imm1295e-page1 too, is it only a reftest issue or is it occurring in the actual viewer too?

Locally: there are no problem... pffff
And it's ok with chrome.
I dumped the generated svg for all the page 1, copy/paste it in a file and then open it in Firefox: no problem at all.

calixteman · 2021-07-28T12:54:26Z

/botio test

pdfjsbot · 2021-07-28T12:54:28Z

From: Bot.io (Windows)

Received

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

Live output at: http://3.101.106.178:8877/7e213a1dc752b54/output.txt

pdfjsbot · 2021-07-28T12:54:28Z

From: Bot.io (Linux m4)

Received

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

Live output at: http://54.67.70.0:8877/6562c837b514ffa/output.txt

pdfjsbot · 2021-07-28T13:27:55Z

From: Bot.io (Linux m4)

Failed

Full output at http://54.67.70.0:8877/6562c837b514ffa/output.txt

Total script time: 33.45 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  different ref/snapshot: 481
  different first/second rendering: 1

Image differences available at: http://54.67.70.0:8877/6562c837b514ffa/reftest-analyzer.html#web=eq.log

pdfjsbot · 2021-07-28T13:33:28Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/7e213a1dc752b54/output.txt

Total script time: 38.99 mins

Font tests: Passed
Unit tests: Passed
Integration Tests: Passed
Regression tests: FAILED

  different ref/snapshot: 479
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/7e213a1dc752b54/reftest-analyzer.html#web=eq.log

marco-c · 2021-07-28T13:34:39Z

@calixteman can you check firefox-hsbc-page1 too? It also has white fields.

calixteman · 2021-07-28T14:33:43Z

@calixteman can you check firefox-hsbc-page1 too? It also has white fields.

The chrome snapshot is ok for this one and locally there are no problems.

calixteman · 2021-07-28T15:01:23Z

/botio makeref

pdfjsbot · 2021-07-28T15:01:24Z

From: Bot.io (Linux m4)

Received

Command cmd_makeref from @calixteman received. Current queue size: 0

Live output at: http://54.67.70.0:8877/d1c5cc9ededc5d6/output.txt

pdfjsbot · 2021-07-28T15:01:24Z

From: Bot.io (Windows)

Received

Command cmd_makeref from @calixteman received. Current queue size: 1

Live output at: http://3.101.106.178:8877/a198e7565cf1c9d/output.txt

pdfjsbot · 2021-07-28T15:31:41Z

From: Bot.io (Linux m4)

Success

Full output at http://54.67.70.0:8877/d1c5cc9ededc5d6/output.txt

Total script time: 30.28 mins

Lint: Passed
Make references: Passed
Check references: Passed

pdfjsbot · 2021-07-28T15:35:32Z

From: Bot.io (Windows)

Failed

Full output at http://3.101.106.178:8877/a198e7565cf1c9d/output.txt

Total script time: 29.09 mins

Lint: Passed
Make references: FAILED

marco-c reviewed Jul 27, 2021

View reviewed changes

src/core/xfa/template.js Outdated Show resolved Hide resolved

calixteman force-pushed the xfa_fields branch from 73d5f4a to 86565cd Compare July 27, 2021 18:04

timvandermeij added the form-xfa label Jul 27, 2021

XFA - Fix auto-sized fields (bug 1722030)

76d882b

- In order to better compute text fields size, use line height with no gaps (and consequently guessed height for text are slightly better in general). - Fix default background color in fields.

calixteman force-pushed the xfa_fields branch from 86565cd to 00a379b Compare July 28, 2021 09:35

calixteman force-pushed the xfa_fields branch from 00a379b to e64078c Compare July 28, 2021 11:54

XFA - Add a transparent blue background on all text fields for consis…

92f4cc5

…tency

calixteman force-pushed the xfa_fields branch from e64078c to 92f4cc5 Compare July 28, 2021 12:48

marco-c approved these changes Jul 28, 2021

View reviewed changes

calixteman merged commit ac5c4b7 into mozilla:master Jul 28, 2021

calixteman mentioned this pull request Jul 29, 2021

XFA Subforms incorrectly colored #13747

Closed

marco-c linked an issue Jul 29, 2021 that may be closed by this pull request

XFA Subforms incorrectly colored #13747

Closed

XFA - Fix auto-sized fields (bug 1722030) #13806

XFA - Fix auto-sized fields (bug 1722030) #13806

Conversation

calixteman commented Jul 27, 2021

calixteman commented Jul 27, 2021

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Failed

marco-c commented Jul 27, 2021

calixteman commented Jul 27, 2021

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Failed

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Failed

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Failed

marco-c commented Jul 27, 2021

calixteman commented Jul 27, 2021

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Received

pdfjsbot commented Jul 27, 2021

From: Bot.io (Windows)

Failed

pdfjsbot commented Jul 27, 2021

From: Bot.io (Linux m4)

Failed

marco-c commented Jul 27, 2021

calixteman commented Jul 28, 2021 • edited Loading

Snuffleupagus commented Jul 28, 2021

marco-c commented Jul 28, 2021

calixteman commented Jul 28, 2021

calixteman commented Jul 28, 2021

pdfjsbot commented Jul 28, 2021

From: Bot.io (Windows)

Received

pdfjsbot commented Jul 28, 2021

From: Bot.io (Linux m4)

Received

pdfjsbot commented Jul 28, 2021

From: Bot.io (Windows)

Failed

pdfjsbot commented Jul 28, 2021

From: Bot.io (Linux m4)

Failed

calixteman commented Jul 28, 2021

calixteman commented Jul 28, 2021

pdfjsbot commented Jul 28, 2021

From: Bot.io (Linux m4)

Received

pdfjsbot commented Jul 28, 2021

From: Bot.io (Windows)

Received

pdfjsbot commented Jul 28, 2021

From: Bot.io (Linux m4)

Failed

marco-c commented Jul 28, 2021

pdfjsbot commented Jul 28, 2021

From: Bot.io (Windows)

Failed

marco-c commented Jul 28, 2021

calixteman commented Jul 28, 2021

calixteman commented Jul 28, 2021 •

edited

Loading