mikesamuel - Project Hosting on Google Code

User Profile		Updates

Today 7 hours ago google-caja issue 1180 (If URICallback.rewrite for a url returns null, the cajoler r...) changed - Yes. We should be able to cajole it. There's no reason to force javascript: URLs through the URI handler just as there's no reason to force same page anchor references through the URI handler -- there's nothing uniform about them. Status: Started Owner: mikesamuel Yes. We should be able to cajole it. There's no reason to force javascript: URLs through the URI handler just as there's no reason to force same page anchor references through the URI handler -- there's nothing uniform about them. Status: Started Owner: mikesamuel Yesterday 28 hours ago google-caja r3916 (Get OpenSocial conformance test suite cajoling without warni...) committed - Get OpenSocial conformance test suite cajoling without warnings http://codereview.appspot.com/179073 This changes our handling of colors so that the X11 color set is recognized and transformed to CSS2.1 compatible hex values. This also changes css-extensions to support the rgba(...) form for colors. With this change, running $ bin/runcaja com.google.caja.opensocial.GadgetRewriterMain \ -i 'http://opensocial- resources.googlecode.com/svn/tests/trunk/suites/0.8/compliance/reference.xml' \ -o /tmp/testsout/reference yields a bunch of semicolon and minor HTML lint messages and LINT: test.css:3+26 - 35: replacing non-standard color cadetblue with hex test.css:3: border-top: 1px dotted cadetblue; ^^^^^^^^^ LINT: test.css:52+17 - 30: replacing non-standard color darkgoldenrod with hex test.css:52: border-color: darkgoldenrod; ^^^^^^^^^^^^^ LINT: test.css:93+21 - 30: replacing non-standard color mistyrose with hex test.css:93: background-color: mistyrose; ^^^^^^^^^ which should make the would-be-interior-designers over in Shindig land happy. R=jasvir@gmail.com Get OpenSocial conformance test suite cajoling without warnings http://codereview.appspot.com/179073 This changes our handling of colors so that the X11 color set is recognized and transformed to CSS2.1 compatible hex values. This also changes css-extensions to support the rgba(...) form for colors. With this change, running $ bin/runcaja com.google.caja.opensocial.GadgetRewriterMain \ -i 'http://opensocial- resources.googlecode.com/svn/tests/trunk/suites/0.8/compliance/reference.xml' \ -o /tmp/testsout/reference yields a bunch of semicolon and minor HTML lint messages and LINT: test.css:3+26 - 35: replacing non-standard color cadetblue with hex test.css:3: border-top: 1px dotted cadetblue; ^^^^^^^^^ LINT: test.css:52+17 - 30: replacing non-standard color darkgoldenrod with hex test.css:52: border-color: darkgoldenrod; ^^^^^^^^^^^^^ LINT: test.css:93+21 - 30: replacing non-standard color mistyrose with hex test.css:93: background-color: mistyrose; ^^^^^^^^^ which should make the would-be-interior-designers over in Shindig land happy. R=jasvir@gmail.com Last 7 days Dec 14, 2009 browserscope r446 (add JSKB tests for function names and prototype chains) committed - add JSKB tests for function names and prototype chains add JSKB tests for function names and prototype chains Dec 14, 2009 google-caja r3914 (Replace htmlparser 1.0.7 with htmlparser 1.2.1 http://codere...) committed - Replace htmlparser 1.0.7 with htmlparser 1.2.1 http://codereview.appspot.com/165082 This modifies htmlparser1.2.1 to expose ElementName and AttributeName. This means that we're parsing an up-to-date version of HTML5, but the new parser delays creating text nodes, which leads to bad file positions. This should be fixable in later CLs, but is a defect of the current implementation. R=jasvir@gmail.com, fargo@google.com Replace htmlparser 1.0.7 with htmlparser 1.2.1 http://codereview.appspot.com/165082 This modifies htmlparser1.2.1 to expose ElementName and AttributeName. This means that we're parsing an up-to-date version of HTML5, but the new parser delays creating text nodes, which leads to bad file positions. This should be fixable in later CLs, but is a defect of the current implementation. R=jasvir@gmail.com, fargo@google.com Dec 14, 2009 google-code-prettify issue 103 (Support for VHDL language) changed - Thanks much. Added at r96 : http://code.google.com/p/google-code-prettify/source/detail?r=96 with one change to make sure that a \ delimited identifier can follow punctuation without intervening space. Status: Fixed Owner: mikesamuel Thanks much. Added at r96 : http://code.google.com/p/google-code-prettify/source/detail?r=96 with one change to make sure that a \ delimited identifier can follow punctuation without intervening space. Status: Fixed Owner: mikesamuel Dec 14, 2009 google-code-prettify r96 (added Le Poussin's VHDL mode) committed - added Le Poussin's VHDL mode added Le Poussin's VHDL mode Last 30 days Dec 10, 2009 google-caja r3911 (Cleanup warnings http://codereview.appspot.com/175043 R=j...) committed - Cleanup warnings http://codereview.appspot.com/175043 R=jasvir@gmail.com Cleanup warnings http://codereview.appspot.com/175043 R=jasvir@gmail.com Dec 10, 2009 google-code-prettify issue 102 (Performance bug with newline replacement on IE 6-7 (patch in...) changed - Patched in at revision 95. Thanks much. Status: Fixed Owner: mikesamuel Patched in at revision 95. Thanks much. Status: Fixed Owner: mikesamuel Dec 10, 2009 google-code-prettify r95 (patched in jacobly's IE performance improvements from issue ...) committed - patched in jacobly's IE performance improvements from issue 102 patched in jacobly's IE performance improvements from issue 102 Dec 10, 2009 google-caja r3910 (Moved CajaWebToolsServlet.main into a separate class so that...) committed - Moved CajaWebToolsServlet.main into a separate class so that it doesn't interact badly with appengine. Moved CajaWebToolsServlet.main into a separate class so that it doesn't interact badly with appengine. Dec 10, 2009 google-caja r3909 (A testcase for bug 1029 http://codereview.appspot.com/171054...) committed - A testcase for bug 1029 http://codereview.appspot.com/171054 Bug 1029 seems to have been fixed, but this testcase increases coverage. R=metaweta@gmail.com A testcase for bug 1029 http://codereview.appspot.com/171054 Bug 1029 seems to have been fixed, but this testcase increases coverage. R=metaweta@gmail.com Dec 10, 2009 google-code-prettify issue 102 (Performance bug with newline replacement on IE 6-7 (patch in...) commented on - Thanks so much for profiling. It's really hard for me to run one since I can only run IE under VMWare which is a royal PITA. I'll try incorporating this when I get a chance. Do you know where time under IE8 is spent? Thanks so much for profiling. It's really hard for me to run one since I can only run IE under VMWare which is a royal PITA. I'll try incorporating this when I get a chance. Do you know where time under IE8 is spent? Dec 10, 2009 google-code-prettify issue 101 (HTML highlighter bugs on multiple <script> tags) Status changed - I tried your example and it seems to work at revision 92. Please reopen this issue if upgrading doesn't solve your problem. Status: Fixed I tried your example and it seems to work at revision 92. Please reopen this issue if upgrading doesn't solve your problem. Status: Fixed Dec 09, 2009 google-caja issue 1029 (Css UriLiterals do not clone properly if URI has no scheme) Status changed - Seems to be working as of revision 3900. Status: Fixed Seems to be working as of revision 3900. Status: Fixed Dec 09, 2009 google-caja issue 1172 (window is undefined in the testbed) commented on - Ok. Hopefully it's applet weirdness that won't bite us once we bury the applet testbed, but let's remember it in case we see it again. Ok. Hopefully it's applet weirdness that won't bite us once we bury the applet testbed, but let's remember it in case we see it again. Dec 09, 2009 google-caja issue 1156 (JSON.parse infinite loop in browsers with native JSON) changed - Is it the case that all these browsers support native JSON, hence that code path is exercised? Status: Accepted Owner: mikesamuel Is it the case that all these browsers support native JSON, hence that code path is exercised? Status: Accepted Owner: mikesamuel Dec 09, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) Status changed - Status: Pending Status: Pending Dec 09, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) commented on - http://codereview.appspot.com/171052/show http://codereview.appspot.com/171052/show Dec 09, 2009 google-caja issue 1165 (domita_test testCustomEvent fails on opera 10) Status changed - Fixed @3899 Status: Fixed Fixed @3899 Status: Fixed Dec 09, 2009 google-caja r3899 (Issue 1165: fixed testCustomEvent on opera http://codereview...) committed - Issue 1165: fixed testCustomEvent on opera http://codereview.appspot.com/167059 Opera supports both addEventListener and attachEvent but events created using the DOM standard mechanism can only be dispatched to listeners added using addEventListener. R=felix8a@gmail.com Issue 1165: fixed testCustomEvent on opera http://codereview.appspot.com/167059 Opera supports both addEventListener and attachEvent but events created using the DOM standard mechanism can only be dispatched to listeners added using addEventListener. R=felix8a@gmail.com Dec 09, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) changed - Ok, so rewriters that are a core part of the shindig project would need to be changed. And possibly some proprietary client specific code that does similar things would also need to be changed. Status: Accepted Owner: mikesamuel Ok, so rewriters that are a core part of the shindig project would need to be changed. And possibly some proprietary client specific code that does similar things would also need to be changed. Status: Accepted Owner: mikesamuel Dec 09, 2009 browserscope r440 (fleshed out docs and shortened test result summaries, and es...) committed - fleshed out docs and shortened test result summaries, and escape code in help text properly. fleshed out docs and shortened test result summaries, and escape code in help text properly. Dec 09, 2009 google-caja issue 1174 (bridal <button> clone in ie[67]) Status changed - Submitted @3898 Status: Fixed Submitted @3898 Status: Fixed Dec 09, 2009 google-caja r3898 (Issue 1174 - <button>.cloneNode fails on IE http://coderevie...) committed - Issue 1174 - <button>.cloneNode fails on IE http://codereview.appspot.com/165098 http://code.google.com/p/google-caja/issues/detail?id=1174 ie[67] doesn't let you change a button's type after it's created. it's the same as the <input> type problem, you have to specify the type in createElement. bridal's cloneNode handles <input> correctly but not <button. This change fixes constructClone in bridal.js. I also checked html4-attributes-defs.json to see whether there are any other nodes with a "type" or "value" attribute that need to be fixed up. The only other ones are PARAM and LI, UK, OL, and disallowed elements like OBJECT and LINK, none of which are inputs. R=felix8a@gmail.com Issue 1174 - <button>.cloneNode fails on IE http://codereview.appspot.com/165098 http://code.google.com/p/google-caja/issues/detail?id=1174 ie[67] doesn't let you change a button's type after it's created. it's the same as the <input> type problem, you have to specify the type in createElement. bridal's cloneNode handles <input> correctly but not <button. This change fixes constructClone in bridal.js. I also checked html4-attributes-defs.json to see whether there are any other nodes with a "type" or "value" attribute that need to be fixed up. The only other ones are PARAM and LI, UK, OL, and disallowed elements like OBJECT and LINK, none of which are inputs. R=felix8a@gmail.com Dec 09, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) commented on - Which rewriters? Which rewriters? Dec 08, 2009 google-caja issue 1165 (domita_test testCustomEvent fails on opera 10) changed - http://codereview.appspot.com/167059 Status: Pending Owner: mikesamuel http://codereview.appspot.com/167059 Status: Pending Owner: mikesamuel Dec 08, 2009 google-caja issue 1176 (Ant build target "demos" fails with build.xml format error) Status changed - Submitted @3897 Status: Fixed Submitted @3897 Status: Fixed Dec 08, 2009 google-caja r3897 (Fix issue 1176: ant demos fails with ANT error http://codere...) committed - Fix issue 1176: ant demos fails with ANT error http://codereview.appspot.com/167054 http://code.google.com/p/google-caja/issues/detail?id=1176 > What steps will reproduce the problem? 1. Checkout using svn (or git-svn) 2. (cd caja && ant demos) > What do you see instead? BUILD FAILED /opt/svn/google-caja/build.xml:1121: transform doesn't support the nested "include" element. R=metaweta@gmail.com Fix issue 1176: ant demos fails with ANT error http://codereview.appspot.com/167054 http://code.google.com/p/google-caja/issues/detail?id=1176 > What steps will reproduce the problem? 1. Checkout using svn (or git-svn) 2. (cd caja && ant demos) > What do you see instead? BUILD FAILED /opt/svn/google-caja/build.xml:1121: transform doesn't support the nested "include" element. R=metaweta@gmail.com Dec 08, 2009 google-caja issue 1172 (window is undefined in the testbed) commented on - Which testbed? The applet based one or the applet-free one? Which testbed? The applet based one or the applet-free one? Dec 08, 2009 google-caja issue 1174 (bridal <button> clone in ie[67]) changed - http://codereview.appspot.com/165098 Status: Pending Owner: mikesamuel http://codereview.appspot.com/165098 Status: Pending Owner: mikesamuel Dec 08, 2009 google-caja issue 1176 (Ant build target "demos" fails with build.xml format error) changed - http://codereview.appspot.com/167054 Status: Pending Owner: mikesamuel http://codereview.appspot.com/167054 Status: Pending Owner: mikesamuel Dec 08, 2009 google-caja issue 1176 (Ant build target "demos" fails with build.xml format error) commented on - This is cruft left over from http://code.google.com/p/google-caja/source/detail?r=3841 This is cruft left over from http://code.google.com/p/google-caja/source/detail?r=3841 Dec 08, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) Cc changed - Ok. So Shindig parses it as a non-namespace aware document. I think Fargo is updating Shindig to use a namespace aware parser. In the meantime, we can probably treat a null namespace as HTML as long as the qualified name contains no :. Cc: johnfargo Ok. So Shindig parses it as a non-namespace aware document. I think Fargo is updating Shindig to use a namespace aware parser. In the meantime, we can probably treat a null namespace as HTML as long as the qualified name contains no :. Cc: johnfargo Dec 08, 2009 browserscope r439 (first pass at JSON serving for JSKB) committed - first pass at JSON serving for JSKB first pass at JSON serving for JSKB Dec 08, 2009 google-caja issue 1177 (NPE in Namespaces.java:131) commented on - NPE is definitely not the best diagnostic. Giles, can you explain how rootNode was created? NPE is definitely not the best diagnostic. Giles, can you explain how rootNode was created? Dec 08, 2009 google-code-prettify issue 101 (HTML highlighter bugs on multiple <script> tags) commented on - Thanks. http://code.google.com/p/google-code-prettify/source/detail?r=92 may have already fixed the issue. I'll check when I have a spare minute. Thanks. http://code.google.com/p/google-code-prettify/source/detail?r=92 may have already fixed the issue. I'll check when I have a spare minute. Dec 08, 2009 google-code-prettify issue 101 (HTML highlighter bugs on multiple <script> tags) commented on - What is the problem you're seeing? What is the problem you're seeing? Dec 08, 2009 browserscope r437 (Fixed bugs that broke JSKB on IE. Array.indexOf not defined...) committed - Fixed bugs that broke JSKB on IE. Array.indexOf not defined on IE. Also fixed bug with JSON array rendering. Fixed bugs that broke JSKB on IE. Array.indexOf not defined on IE. Also fixed bug with JSON array rendering. Dec 07, 2009 browserscope r434 (fixed typo with parentheses) committed - fixed typo with parentheses fixed typo with parentheses Dec 07, 2009 browserscope r425 (stub out json serving) committed - stub out json serving stub out json serving Dec 07, 2009 json-sans-eval r12 (fixed comment to work with closure compiler) committed - fixed comment to work with closure compiler fixed comment to work with closure compiler Dec 05, 2009 browserscope r424 (fix scoring, flesh out test descriptions, and add test for w...) committed - fix scoring, flesh out test descriptions, and add test for whether finally works fix scoring, flesh out test descriptions, and add test for whether finally works Dec 04, 2009 google-caja FinallySkipped (finally blocks can fail to execute in one script block and c...) Wiki page edited - Revision r3895 Edited wiki page through web user interface. Revision r3895 Edited wiki page through web user interface. Dec 04, 2009 google-caja FinallySkipped ({{{finally}}} blocks can fail to execute in one script block...) Wiki page added - Revision r3894 Created wiki page through web user interface. Revision r3894 Created wiki page through web user interface. Dec 04, 2009 google-caja AttackVectors (Interpreter&Browser properties that can be exploited to esca...) Wiki page edited - Revision r3893 Edited wiki page through web user interface. Revision r3893 Edited wiki page through web user interface. Dec 04, 2009 google-caja r3892 (Unpack HTML Parser source into third_party/java/htmlparser h...) committed - Unpack HTML Parser source into third_party/java/htmlparser http://codereview.appspot.com/166051 This is a prerequisite to being able to modify the source to let us upgrade to a modern version of libhtmlparser. This is the unmodified source for revision 1.2.1. R=jasvir Unpack HTML Parser source into third_party/java/htmlparser http://codereview.appspot.com/166051 This is a prerequisite to being able to modify the source to let us upgrade to a modern version of libhtmlparser. This is the unmodified source for revision 1.2.1. R=jasvir Dec 04, 2009 google-caja r3891 (Fix toInt32 and toUint32 around NaN and Infinity http://code...) committed - Fix toInt32 and toUint32 around NaN and Infinity http://codereview.appspot.com/157109 Operation.fold depended on a faulty implementation of toInt32 and toUint32. R=davidsarah.hopwood@gmail.com, metaweta@gmail.com Fix toInt32 and toUint32 around NaN and Infinity http://codereview.appspot.com/157109 Operation.fold depended on a faulty implementation of toInt32 and toUint32. R=davidsarah.hopwood@gmail.com, metaweta@gmail.com Dec 04, 2009 google-code-prettify issue 100 (Support for "<code>" tag) Status changed - The code in the third bullet point: 3. Add onload="prettyPrint()" to your document's body tag. at http://google-code-prettify.googlecode.com/svn/trunk/README.html is in a <code> element. You never need to use a lang-* class, you only need to use the prettyprint class. If it doesn't work for you after following those steps, please reopen this bug and tell me what browser you're using. A link to sample code that shows the bug would be great too. Status: Invalid The code in the third bullet point: 3. Add onload="prettyPrint()" to your document's body tag. at http://google-code-prettify.googlecode.com/svn/trunk/README.html is in a <code> element. You never need to use a lang-* class, you only need to use the prettyprint class. If it doesn't work for you after following those steps, please reopen this bug and tell me what browser you're using. A link to sample code that shows the bug would be great too. Status: Invalid Dec 03, 2009 google-caja r3888 (warning cleanup in ancillary tools) committed - warning cleanup in ancillary tools warning cleanup in ancillary tools Dec 03, 2009 google-caja r3887 (fixed optimizer to recurse into return values, and const loc...) committed - fixed optimizer to recurse into return values, and const local optimizer to be careful about inlining long strings fixed optimizer to recurse into return values, and const local optimizer to be careful about inlining long strings

r3916 (Get OpenSocial conformance test suite cajoling without warni...) committed - Get OpenSocial conformance test suite cajoling without warnings http://codereview.appspot.com/179073 This changes our handling of colors so that the X11 color set is recognized and transformed to CSS2.1 compatible hex values. This also changes css-extensions to support the rgba(...) form for colors. With this change, running $ bin/runcaja com.google.caja.opensocial.GadgetRewriterMain \ -i 'http://opensocial- resources.googlecode.com/svn/tests/trunk/suites/0.8/compliance/reference.xml' \ -o /tmp/testsout/reference yields a bunch of semicolon and minor HTML lint messages and LINT: test.css:3+26 - 35: replacing non-standard color cadetblue with hex test.css:3: border-top: 1px dotted cadetblue; ^^^^^^^^^ LINT: test.css:52+17 - 30: replacing non-standard color darkgoldenrod with hex test.css:52: border-color: darkgoldenrod; ^^^^^^^^^^^^^ LINT: test.css:93+21 - 30: replacing non-standard color mistyrose with hex test.css:93: background-color: mistyrose; ^^^^^^^^^ which should make the would-be-interior-designers over in Shindig land happy. R=jasvir@gmail.com

Get OpenSocial conformance test suite cajoling without warnings http://codereview.appspot.com/179073 This changes our handling of colors so that the X11 color set is recognized and transformed to CSS2.1 compatible hex values. This also changes css-extensions to support the rgba(...) form for colors. With this change, running $ bin/runcaja com.google.caja.opensocial.GadgetRewriterMain \ -i 'http://opensocial- resources.googlecode.com/svn/tests/trunk/suites/0.8/compliance/reference.xml' \ -o /tmp/testsout/reference yields a bunch of semicolon and minor HTML lint messages and LINT: test.css:3+26 - 35: replacing non-standard color cadetblue with hex test.css:3: border-top: 1px dotted cadetblue; ^^^^^^^^^ LINT: test.css:52+17 - 30: replacing non-standard color darkgoldenrod with hex test.css:52: border-color: darkgoldenrod; ^^^^^^^^^^^^^ LINT: test.css:93+21 - 30: replacing non-standard color mistyrose with hex test.css:93: background-color: mistyrose; ^^^^^^^^^ which should make the would-be-interior-designers over in Shindig land happy. R=jasvir@gmail.com

Today

Yesterday

Last 7 days

Last 30 days