[MediaWiki-commits] [Gerrit] Refactor wts to emphasize cb is an accumulator - change (mediawiki...parsoid)

jenkins-bot (Code Review) Tue, 03 Nov 2015 10:45:02 -0800

jenkins-bot has submitted this change and it was merged.

Change subject: Refactor wts to emphasize cb is an accumulator
......................................................................



Refactor wts to emphasize cb is an accumulator

 * The passing test is from preserving `atStartOfOutput` after the
   serializeToString use.

Change-Id: I8e3b6c60c97d58f63b0d52e030b4c51055060bc6
---
M lib/html2wt/ConstrainedText.js
M lib/html2wt/SerializerState.js
M lib/html2wt/WikitextSerializer.js
M tests/parserTests-blacklist.js
4 files changed, 33 insertions(+), 34 deletions(-)

Approvals:
  Subramanya Sastry: Looks good to me, approved
  jenkins-bot: Verified



diff --git a/lib/html2wt/ConstrainedText.js b/lib/html2wt/ConstrainedText.js
index 85cc33e..1068755 100644
--- a/lib/html2wt/ConstrainedText.js
+++ b/lib/html2wt/ConstrainedText.js
@@ -27,7 +27,7 @@
  * processing).
  * @private
  */
-var escapeLine = function(line, cb) {
+var escapeLine = function(line) {
        // The left context will be precise (that is, it is the result
        // of `ConstrainedText#escape` and will include any escapes
        // triggered by chunks on the left), but the right context
@@ -51,12 +51,8 @@
                // Process the escapes for this chunk, given escaped previous 
chunk
                state.rightContext = 
state.rightContext.slice(chunk.text.length);
                var thisEscape = chunk.escape(state);
-               var origPos = state.leftContext.length;
                state.leftContext +=
                        (thisEscape.prefix || '') + thisEscape.text + 
(thisEscape.suffix || '');
-               if (cb) {
-                       cb(state.leftContext.slice(origPos), chunk.node);
-               }
                if (thisEscape.greedy) {
                        // protect the left context: this will be matched 
greedily
                        // by this chunk, so there's no chance that a subsequent
diff --git a/lib/html2wt/SerializerState.js b/lib/html2wt/SerializerState.js
index ee69f37..c9d2f54 100644
--- a/lib/html2wt/SerializerState.js
+++ b/lib/html2wt/SerializerState.js
@@ -132,6 +132,7 @@
        Util.extendProps(this, Util.clone(options), Util.clone(initialState));
        this.resetCurrLine(null);
        this.singleLineContext = new SingleLineContext();
+       this.out = '';
 }
 
 var SSP = SerializerState.prototype;
@@ -151,8 +152,8 @@
 
 /**
  */
-SSP.flushLine = function(cb) {
-       escapeLine(this.currLine.chunks, cb);
+SSP.flushLine = function() {
+       this.out += escapeLine(this.currLine.chunks);
        this.currLine.chunks.length = 0;
 };
 
@@ -244,7 +245,7 @@
 
 /**
  */
-SSP.emitSepAndOutput = function(res, node, cb, logPrefix) {
+SSP.emitSepAndOutput = function(res, node, logPrefix) {
        res = ConstrainedText.cast(res, node);
 
        // Replace newlines if we're in a single-line context
@@ -327,7 +328,7 @@
 
        if (this.onSOL) {
                // process escapes in our full line
-               this.flushLine(cb);
+               this.flushLine();
                this.resetCurrLine(node);
        }
 
@@ -432,6 +433,9 @@
        if (!solRE.test(res)) {
                this.onSOL = false;
        }
+
+       // We've emit something so we're no longer at SOO.
+       this.atStartOfOutput = false;
 };
 
 /**
@@ -441,29 +445,36 @@
 SSP._serializeChildrenToString = function(node, wtEscaper, inState) {
        // FIXME: Make sure that the separators emitted here conform to the
        // syntactic constraints of syntactic context.
-       var bits = '';
        var oldSep = this.sep;
        var oldSOL = this.onSOL;
+       var oldOut = this.out;
+       var oldStart = this.atStartOfOutput;
        var oldChunks = this.currLine.chunks;
-       // appendToBits just ignores anything returned but
-       // the source, but that is fine. Selser etc is handled in
-       // the top level callback at a slightly coarser level.
-       var appendToBits = function(out) { bits += out; };
        var self = this;
        var cb = function(res, node) {
-               self.emitSepAndOutput(res, node, appendToBits, "OUT(C):");
+               self.emitSepAndOutput(res, node, "OUT(C):");
        };
+
+       this.out = '';
        this.sep = {};
        this.onSOL = false;
+       this.atStartOfOutput = false;
        this.currLine.chunks = [];
        this[inState] = true;
+
        this.serializeChildren(node, cb, wtEscaper);
-       this.flushLine(appendToBits);
-       self.serializer.buildAndEmitSep(this, appendToBits, node);
+       // Emit child-parent seps,
+       cb('', node);
+       // We've reached EOF, flush the remaining buffered text.
+       this.flushLine();
+
        // restore the state
+       var bits = this.out;
+       this.out = oldOut;
        this[inState] = false;
        this.sep = oldSep;
        this.onSOL = oldSOL;
+       this.atStartOfOutput = oldStart;
        this.currLine.chunks = oldChunks;
        return bits;
 };
diff --git a/lib/html2wt/WikitextSerializer.js 
b/lib/html2wt/WikitextSerializer.js
index 5905c6e..cbb345e 100644
--- a/lib/html2wt/WikitextSerializer.js
+++ b/lib/html2wt/WikitextSerializer.js
@@ -1366,34 +1366,27 @@
                return "--- DOM --- \n" + body.outerHTML + "\n-----------";
        });
 
-       var out = '';
-
        state.sep.lastSourceNode = body;
        state.currLine.firstNode = body.firstChild;
 
        // Wrapper CB for every chunk that emits any required separators
        // before emitting the chunk itself.
-       var chunkCBWrapper = function(chunk, node, atEOF) {
-               var accum = function(o) { out += o; };
-               state.emitSepAndOutput(chunk, node, accum, "OUT:");
-               state.atStartOfOutput = false;
-               if (atEOF === 'EOF') {
-                       state.flushLine(accum);
-               }
+       var chunkCBWrapper = function(chunk, node) {
+               state.emitSepAndOutput(chunk, node, "OUT:");
        };
 
-       // Kick it off
        state.serializeChildren(body, chunkCBWrapper);
-
-       // Handle EOF
-       chunkCBWrapper('', body, 'EOF');
+       // Emit child-parent seps.
+       chunkCBWrapper('', body);
+       // We've reached EOF, flush the remaining buffered text.
+       state.flushLine();
 
        if (state.hasIndentPreNowikis) {
                // FIXME: Perhaps this can be done on a per-line basis
                // rather than do one post-pass on the entire document.
                //
                // Strip excess/useless nowikis
-               out = stripUnnecessaryIndentPreNowikis(this.env, out);
+               state.out = stripUnnecessaryIndentPreNowikis(this.env, 
state.out);
        }
 
        if (state.hasQuoteNowikis) {
@@ -1401,10 +1394,10 @@
                // rather than do one post-pass on the entire document.
                //
                // Strip excess/useless nowikis
-               out = stripUnnecessaryQuoteNowikis(out);
+               state.out = stripUnnecessaryQuoteNowikis(state.out);
        }
 
-       return out;
+       return state.out;
 };
 
 
diff --git a/tests/parserTests-blacklist.js b/tests/parserTests-blacklist.js
index 2fee96d..30c9a80 100644
--- a/tests/parserTests-blacklist.js
+++ b/tests/parserTests-blacklist.js
@@ -526,7 +526,6 @@
 add("html2html", "Link to image page- image page normally doesn't exists, 
hence edit link\nAdd test with existing image page\n#<p><a 
href=\"/wiki/File:Test\" title=\"Image:Test\">Image:test</a>", "<p 
data-parsoid='{\"dsr\":[0,61,0,0]}'>[/index.php?title=File:Test&amp;action=edit&amp;redlink=1
 Image:test]</p>\n");
 add("html2html", "bug 18784  Link to non-existent image page with caption 
should use caption as link text", "<p 
data-parsoid='{\"dsr\":[0,58,0,0]}'>[/index.php?title=File:Test&amp;action=edit&amp;redlink=1
 caption]</p>\n");
 add("html2html", "SVG thumbnails with invalid language code", "<figure 
class=\"mw-default-size\" typeof=\"mw:Image/Thumb\" 
data-parsoid='{\"optList\":[{\"ck\":\"thumbnail\",\"ak\":\"thumb\"},{\"ck\":\"lang\",\"ak\":\"lang=invalid.language.code\"}],\"dsr\":[0,52,2,2]}'><a
 href=\"./File:Foobar.svg\" 
data-parsoid='{\"a\":{\"href\":\"./File:Foobar.svg\"},\"sa\":{},\"dsr\":[2,50,null,null]}'><img
 resource=\"./File:Foobar.svg\" 
src=\"//example.com/images/thumb/f/ff/Foobar.svg/220px-Foobar.svg\" 
lang=\"invalid.language.code\" data-file-width=\"240\" data-file-height=\"180\" 
data-file-type=\"drawing\" height=\"165\" width=\"220\" 
data-parsoid='{\"a\":{\"resource\":\"./File:Foobar.svg\",\"lang\":\"invalid.language.code\",\"height\":\"165\",\"width\":\"220\"},\"sa\":{\"resource\":\"File:Foobar.svg\",\"lang\":\"lang=invalid.language.code\"}}'/></a></figure>\n");
-add("html2html", "Image: caption containing a table", "<figure 
typeof=\"mw:Image/Thumb\" 
data-parsoid='{\"optList\":[{\"ck\":\"thumbnail\",\"ak\":\"thumb\"},{\"ck\":\"width\",\"ak\":\"200x200px\"},{\"ck\":\"bogus\",\"ak\":\"This
 is an example image thumbnail caption with a 
table\\n{\"},{\"ck\":\"bogus\",\"ak\":\"\\n\\n!Foo 
!Bar\\n\"},{\"ck\":\"bogus\",\"ak\":\"-\\n\"},{\"ck\":\"bogus\",\"ak\":\"Foo1 
\\n\"},{\"ck\":\"bogus\",\"ak\":\"Bar1\"},{\"ck\":\"caption\",\"ak\":\"}\\nand 
some more text.\"}],\"dsr\":[0,143,2,2]}'><a href=\"./File:Foobar.jpg\" 
data-parsoid='{\"a\":{\"href\":\"./File:Foobar.jpg\"},\"sa\":{},\"dsr\":[2,null,null,null]}'><img
 resource=\"./File:Foobar.jpg\" 
src=\"//example.com/images/thumb/3/3a/Foobar.jpg/200px-Foobar.jpg\" 
data-file-width=\"1941\" data-file-height=\"220\" data-file-type=\"bitmap\" 
height=\"23\" width=\"200\" 
data-parsoid='{\"a\":{\"resource\":\"./File:Foobar.jpg\",\"height\":\"23\",\"width\":\"200\"},\"sa\":{\"resource\":\"File:Foobar.jpg\"}}'/></a><figcaption
 data-parsoid='{\"dsr\":[null,141,null,null]}'>}\nand some more 
text.</figcaption></figure>\n");
 add("html2html", "T93580: 3. Templated <ref> inside inline images", "<p 
data-parsoid='{\"dsr\":[0,98,0,0]}'><span class=\"mw-default-size\" 
typeof=\"mw:Image\" 
data-parsoid='{\"optList\":[{\"ck\":\"caption\",\"ak\":\"Undisplayed caption in 
inline image with ref: 
{{echo|&lt;ref>{{echo|foo}}&lt;/ref>}}\"}],\"dsr\":[0,98,null,null]}' 
data-mw='{\"caption\":\"Undisplayed caption in inline image with ref: &lt;span 
about=\\\"#mwt3\\\" class=\\\"mw-ref\\\" id=\\\"cite_ref-1\\\" 
rel=\\\"dc:references\\\" typeof=\\\"mw:Transclusion  mw:Extension/ref\\\" 
data-parsoid=\\\"{&amp;quot;dsr&amp;quot;:[64,96,null,null],&amp;quot;pi&amp;quot;:[[{&amp;quot;k&amp;quot;:&amp;quot;1&amp;quot;,&amp;quot;spc&amp;quot;:[&amp;quot;&amp;quot;,&amp;quot;&amp;quot;,&amp;quot;&amp;quot;,&amp;quot;&amp;quot;]}]]}\\\"
 
data-mw=\\\"{&amp;quot;parts&amp;quot;:[{&amp;quot;template&amp;quot;:{&amp;quot;target&amp;quot;:{&amp;quot;wt&amp;quot;:&amp;quot;echo&amp;quot;,&amp;quot;href&amp;quot;:&amp;quot;./Template:Echo&amp;quot;},&amp;quot;params&amp;quot;:{&amp;quot;1&amp;quot;:{&amp;quot;wt&amp;quot;:&amp;quot;&lt;ref>{{echo|foo}}&lt;/ref>&amp;quot;}},&amp;quot;i&amp;quot;:0}}]}\\\">&lt;a
 href=\\\"#cite_note-1\\\" style=\\\"counter-reset: mw-Ref 1;\\\">&lt;span 
class=\\\"mw-reflink-text\\\">[1]&lt;/span>&lt;/a>&lt;/span>&lt;meta 
typeof=\\\"mw:Transclusion mw:Extension/ref/Marker\\\" about=\\\"#mwt3\\\" 
data-parsoid=\\\"{&amp;quot;group&amp;quot;:&amp;quot;&amp;quot;,&amp;quot;name&amp;quot;:&amp;quot;&amp;quot;,&amp;quot;content&amp;quot;:&amp;quot;foo&amp;quot;,&amp;quot;hasRefInRef&amp;quot;:false,&amp;quot;dsr&amp;quot;:[64,96,null,null],&amp;quot;pi&amp;quot;:[[{&amp;quot;k&amp;quot;:&amp;quot;1&amp;quot;,&amp;quot;spc&amp;quot;:[&amp;quot;&amp;quot;,&amp;quot;&amp;quot;,&amp;quot;&amp;quot;,&amp;quot;&amp;quot;]}]],&amp;quot;tmp&amp;quot;:{}}\\\"
 
data-mw=\\\"{&amp;quot;parts&amp;quot;:[{&amp;quot;template&amp;quot;:{&amp;quot;target&amp;quot;:{&amp;quot;wt&amp;quot;:&amp;quot;echo&amp;quot;,&amp;quot;href&amp;quot;:&amp;quot;./Template:Echo&amp;quot;},&amp;quot;params&amp;quot;:{&amp;quot;1&amp;quot;:{&amp;quot;wt&amp;quot;:&amp;quot;&lt;ref>{{echo|foo}}&lt;/ref>&amp;quot;}},&amp;quot;i&amp;quot;:0}}]}\\\">\"}'><a
 href=\"./File:Foobar.jpg\" 
data-parsoid='{\"a\":{\"href\":\"./File:Foobar.jpg\"},\"sa\":{}}'><img 
resource=\"./File:Foobar.jpg\" src=\"//example.com/images/3/3a/Foobar.jpg\" 
data-file-width=\"1941\" data-file-height=\"220\" data-file-type=\"bitmap\" 
height=\"220\" width=\"1941\" 
data-parsoid='{\"a\":{\"resource\":\"./File:Foobar.jpg\",\"height\":\"220\",\"width\":\"1941\"},\"sa\":{\"resource\":\"File:Foobar.jpg\"}}'/></a></span></p>\n\n<ol
 class=\"mw-references\" typeof=\"mw:Extension/references\" about=\"#mwt7\" 
data-parsoid='{\"dsr\":[100,114,2,2]}' 
data-mw='{\"name\":\"references\",\"attrs\":{}}'><li about=\"#cite_note-1\" 
id=\"cite_note-1\"><a href=\"#cite_ref-1\" rel=\"mw:referencedBy\"><span 
class=\"mw-linkback-text\">↑ </span></a> <span 
id=\"mw-reference-text-cite_note-1\" class=\"mw-reference-text\" 
data-parsoid=\"{}\">foo</span></li></ol>");
 add("html2html", "Subpage link", "<p data-parsoid='{\"dsr\":[0,38,0,0]}'><a 
rel=\"mw:WikiLink\" href=\"./Wiki/Subpage_test/subpage\" title=\"Wiki/Subpage 
test/subpage\" 
data-parsoid='{\"stx\":\"piped\",\"a\":{\"href\":\"./Wiki/Subpage_test/subpage\"},\"sa\":{\"href\":\"wiki/Subpage
 test/subpage\"},\"dsr\":[0,38,28,2]}'>/subpage</a></p>\n");
 add("html2html", "Subpage noslash link", "<p 
data-parsoid='{\"dsr\":[0,37,0,0]}'><a rel=\"mw:WikiLink\" 
href=\"./Wiki/Subpage_test/subpage\" title=\"Wiki/Subpage test/subpage\" 
data-parsoid='{\"stx\":\"piped\",\"a\":{\"href\":\"./Wiki/Subpage_test/subpage\"},\"sa\":{\"href\":\"wiki/Subpage
 test/subpage\"},\"dsr\":[0,37,28,2]}'>subpage</a></p>\n");

-- 
To view, visit https://gerrit.wikimedia.org/r/250603
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings

Gerrit-MessageType: merged
Gerrit-Change-Id: I8e3b6c60c97d58f63b0d52e030b4c51055060bc6
Gerrit-PatchSet: 5
Gerrit-Project: mediawiki/services/parsoid
Gerrit-Branch: master
Gerrit-Owner: Arlolra <[email protected]>
Gerrit-Reviewer: Arlolra <[email protected]>
Gerrit-Reviewer: Cscott <[email protected]>
Gerrit-Reviewer: Subramanya Sastry <[email protected]>
Gerrit-Reviewer: Tim Starling <[email protected]>
Gerrit-Reviewer: jenkins-bot <>

_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits

[MediaWiki-commits] [Gerrit] Refactor wts to emphasize cb is an accumulator - change (mediawiki...parsoid)

Reply via email to