160465 2016-08-02 14:40:18 -0700 Web Inspector: CodeMirror in resource content view should not show 'CR' characters 2016-09-06 16:05:29 -0700 1 1 1 Unclassified WebKit Web Inspector WebKit Nightly Build All All RESOLVED FIXED http://www.tractorsupply.com/tsc/catalog/home-garden/outdoor-power-equipment/generators?cm_re=20160725-_-FOURTH-_-Portable+Generators InRadar P2 Normal --- 1 bburg hi bburg commit-queue hi joepeck mattbaker nvasilyev simon.fraser timothy webkit-bug-importer oldest_to_newest 1216604 0 bburg 2016-08-02 14:40:18 -0700 SUMMARY: When pretty-printing is disabled, or for comment blocks (which aren't altered by pretty-printing), some pages show a 'CR' glyph at the end of each line. This is probably a CR/LF character. We shouldn't render this glyph as it's distracting and irrelevant for JavaScript. STEPS TO REPRODUCE: 1. Go to the URL 2. Go to resources tab 3. Select 'foresee_surveydef.js' or any other JS file served from 'gateway.answerscloud.com' 4. Deselect pretty printing (the {} button) Now most of the lines have a little 'CR' glyph at the end. This is annoying. 1216606 1 webkit-bug-importer 2016-08-02 14:40:48 -0700 <rdar://problem/27666053> 1216624 2 nvasilyev 2016-08-02 15:08:22 -0700 https://github.com/WebKit/webkit/blob/c20f470adf2cf34821a0756d01a4df33b353c1e6/Source/WebInspectorUI/UserInterface/External/CodeMirror/codemirror.js#L6951 This appears to be hardcoded in CodeMirror, i.e. CodeMirror doesn't seem to provide an option to turn it off. We may have to modify codemirror.js directly. 1216634 3 bburg 2016-08-02 15:22:31 -0700 I don't think editing codemirror.js is going to be the right fix. Here are some alternatives: 1. Heuristically detect whether the line endings seem to be \r\n or \n and tell CodeMirror which line ending to assume (there's a setting for that) 2. Strip line endings from the text that we display (somewhat dangerous due to line:col interactions with JSC) 3. Tag the CR character with CSS and make it visibility:none. 1216640 4 nvasilyev 2016-08-02 15:39:33 -0700 (In reply to comment #3) > I don't think editing codemirror.js is going to be the right fix. Here are > some alternatives: > > 1. Heuristically detect whether the line endings seem to be \r\n or \n and > tell CodeMirror which line ending to assume (there's a setting for that) This is what Chrome seems to do. https://codereview.chromium.org/19540026/ > 2. Strip line endings from the text that we display (somewhat dangerous due > to line:col interactions with JSC) > 3. Tag the CR character with CSS and make it visibility:none. I can't figure out how to write a valid CSS selector for this element: <span class="cm-invalidchar" cm-text=" ">␍</span> These don't work: .cm-invalidchar[cm-text=" "] .cm-invalidchar[cm-text="\n"] .cm-invalidchar[cm-text="\a"] 1217951 5 timothy 2016-08-05 10:17:03 -0700 We could just hide all .cm-invalidchar. This should be a setting switch once we have a Settings view. 1226521 6 287986 hi 2016-09-05 16:44:16 -0700 Created attachment 287986 Patch 1226769 7 287986 bburg 2016-09-06 11:40:04 -0700 Comment on attachment 287986 Patch How does this look on misencoded files? Is there any way to differentiate CR from other invalid characters? 1226786 8 hi 2016-09-06 12:22:21 -0700 (In reply to comment #7) > How does this look on misencoded files? Is there any way to differentiate CR > from other invalid characters? Not really sure if I tested that. Any sure-fire steps for reproducing a mis-encoded file? What would that even look like? 1226890 9 287986 bburg 2016-09-06 15:50:22 -0700 Comment on attachment 287986 Patch r=me 1226891 10 bburg 2016-09-06 15:51:01 -0700 (In reply to comment #8) > (In reply to comment #7) > > How does this look on misencoded files? Is there any way to differentiate CR > > from other invalid characters? > > Not really sure if I tested that. Any sure-fire steps for reproducing a > mis-encoded file? What would that even look like? On second thought, I've only really seen this on XHR responses that had the wrong MIME type. I think Johan fixed a bug in that area. Not sure this would really make it any worse... 1226904 11 287986 commit-queue 2016-09-06 16:05:25 -0700 Comment on attachment 287986 Patch Clearing flags on attachment: 287986 Committed r205517: <http://trac.webkit.org/changeset/205517> 1226905 12 commit-queue 2016-09-06 16:05:29 -0700 All reviewed patches have been landed. Closing bug. 287986 2016-09-05 16:44:16 -0700 2016-09-06 16:05:25 -0700 Patch bug-160465-20160905164155.patch text/plain 1363 hi ZGlmZiAtLWdpdCBhL1NvdXJjZS9XZWJJbnNwZWN0b3JVSS9DaGFuZ2VMb2cgYi9Tb3VyY2UvV2Vi SW5zcGVjdG9yVUkvQ2hhbmdlTG9nCmluZGV4IDdhMDZmMDcxYTk5YjM1N2M0NjY0NGRjMWRhYWQx YWM5Y2IxN2EwYTYuLjk4Y2Y5YzI1ZmU4MmUzNzlkOWNiMDFhNjJhYzg1ZDkzZjIxZDNmYmUgMTAw NjQ0Ci0tLSBhL1NvdXJjZS9XZWJJbnNwZWN0b3JVSS9DaGFuZ2VMb2cKKysrIGIvU291cmNlL1dl Ykluc3BlY3RvclVJL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE0IEBACisyMDE2LTA5LTA1ICBEZXZp biBSb3Vzc28gIDxkY3JvdXNzbyt3ZWJraXRAZ21haWwuY29tPgorCisgICAgICAgIFdlYiBJbnNw ZWN0b3I6IENvZGVNaXJyb3IgaW4gcmVzb3VyY2UgY29udGVudCB2aWV3IHNob3VsZCBub3Qgc2hv dyAnQ1InIGNoYXJhY3RlcnMKKyAgICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19i dWcuY2dpP2lkPTE2MDQ2NQorCisgICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgor CisgICAgICAgICogVXNlckludGVyZmFjZS9WaWV3cy9Db2RlTWlycm9yT3ZlcnJpZGVzLmNzczoK KyAgICAgICAgKC5Db2RlTWlycm9yIC5jbS1pbnZhbGlkY2hhcik6CisgICAgICAgIEhpZGUgYWxs IGludmFsaWQgY2hhcmFjdGVycy4KKwogMjAxNi0wOS0wNCAgSm9zZXBoIFBlY29yYXJvICA8cGVj b3Jhcm9AYXBwbGUuY29tPgogCiAgICAgICAgIFdlYiBJbnNwZWN0b3I6IEVuYWJsZSBhbmQgZW5m b3JjZSBzb21lIHJlY29tbWVuZGVkIGFuZCBzdHlsaXN0aWMgRVNMaW50IHJ1bGVzCmRpZmYgLS1n aXQgYS9Tb3VyY2UvV2ViSW5zcGVjdG9yVUkvVXNlckludGVyZmFjZS9WaWV3cy9Db2RlTWlycm9y T3ZlcnJpZGVzLmNzcyBiL1NvdXJjZS9XZWJJbnNwZWN0b3JVSS9Vc2VySW50ZXJmYWNlL1ZpZXdz L0NvZGVNaXJyb3JPdmVycmlkZXMuY3NzCmluZGV4IGY0NjUwMzU1ZTA0ZjcxY2RhZDYxZGZhYWFl M2M3YjkxOTc2YzA4NGUuLmZkNDQ5NDBlMTMyMjk5NmQyMDNlNjQ1YWI0YWMyNWE1N2M2ZDJhMDQg MTAwNjQ0Ci0tLSBhL1NvdXJjZS9XZWJJbnNwZWN0b3JVSS9Vc2VySW50ZXJmYWNlL1ZpZXdzL0Nv ZGVNaXJyb3JPdmVycmlkZXMuY3NzCisrKyBiL1NvdXJjZS9XZWJJbnNwZWN0b3JVSS9Vc2VySW50 ZXJmYWNlL1ZpZXdzL0NvZGVNaXJyb3JPdmVycmlkZXMuY3NzCkBAIC05MCwzICs5MCw4IEBACiAg ICAgY29sb3I6IGhzbCgwLCAwJSwgNjUlKTsKICAgICB0ZXh0LWluZGVudDogMDsKIH0KKworLyog RklYTUU6IEFkZCBzZXR0aW5nIHRvIHNob3cvaGlkZSBpbnZhbGlkIGNoYXJhY3RlcnMuICovCisu Q29kZU1pcnJvciAuY20taW52YWxpZGNoYXIgeworICAgIGRpc3BsYXk6IG5vbmU7Cit9Cg==