-
Notifications
You must be signed in to change notification settings - Fork 2
/
issues-googlecode.xml
38 lines (24 loc) · 33.8 KB
/
issues-googlecode.xml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
<?xml version='1.0' encoding='UTF-8'?><feed xmlns='http://www.w3.org/2005/Atom' xmlns:openSearch='http://a9.com/-/spec/opensearch/1.1/' xmlns:gd='http://schemas.google.com/g/2005' xmlns:issues='http://schemas.google.com/projecthosting/issues/2009'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full</id><updated>2012-01-07T21:18:13.647Z</updated><title>Issues - hyperlink-constituency</title><subtitle>Issues - hyperlink-constituency</subtitle><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/list'/><link rel='http://schemas.google.com/g/2005#feed' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full'/><link rel='http://schemas.google.com/g/2005#post' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full?max-results=22'/><generator version='1.0' uri='http://code.google.com/feeds/issues'>ProjectHosting</generator><openSearch:totalResults>22</openSearch:totalResults><openSearch:startIndex>1</openSearch:startIndex><openSearch:itemsPerPage>22</openSearch:itemsPerPage><entry gd:etag='W/"A0QHR347eCl7ImA9Wx5REU4."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/1</id><published>2010-08-06T14:21:10.000Z</published><updated>2010-08-18T13:48:56.000Z</updated><title>Find dominating node for non-constituents</title><content type='html'>-</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/1/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=1'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/1'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:blocking><issues:id>3</issues:id></issues:blocking><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2010-08-18T13:48:56.000Z</issues:closedDate><issues:id>1</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/117246088971466042425/</issues:uri><issues:username>[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"D04CQX47eCl7ImA9Wx5SFU4."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/2</id><published>2010-08-06T14:24:41.000Z</published><updated>2010-08-11T14:12:40.000Z</updated><title>links db: split out results into multiple columns</title><content type='html'>- main result code
- node label (or dominating node label)
- &quot;almost&quot;, &quot;removed punctuation&quot;, &quot;second pass&quot; flags
- missing node(s) label</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/2/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=2'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/2'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:blocking><issues:id>3</issues:id></issues:blocking><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2010-08-11T14:12:40.000Z</issues:closedDate><issues:id>2</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/117246088971466042425/</issues:uri><issues:username>[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"A0QBSH47eCl7ImA9Wx5REU4."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/3</id><published>2010-08-06T14:25:17.000Z</published><updated>2010-08-18T13:49:19.000Z</updated><title>Record node labels for everything</title><content type='html'>-</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/3/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=3'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/3'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:blockedOn><issues:id>1</issues:id></issues:blockedOn><issues:blockedOn><issues:id>2</issues:id></issues:blockedOn><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2010-08-18T13:49:19.000Z</issues:closedDate><issues:id>3</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/117246088971466042425/</issues:uri><issues:username>[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"DE8AQn47eCl7ImA9Wx5RFEU."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/4</id><published>2010-08-06T14:26:26.000Z</published><updated>2010-08-22T14:20:43.000Z</updated><title>Add LINK and ENDLINK</title><content type='html'>and insert back into the database.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/4/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=4'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/4'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:blocking><issues:id>5</issues:id></issues:blocking><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2010-08-22T14:20:43.000Z</issues:closedDate><issues:id>4</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/117246088971466042425/</issues:uri><issues:username>[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"C0cDQH47eCl7ImA9Wx9XGE0."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/5</id><published>2010-08-06T14:27:08.000Z</published><updated>2011-01-12T02:57:51.000Z</updated><title>Create link/node structure search system</title><content type='html'>-</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/5/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=5'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/5'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:blockedOn><issues:id>4</issues:id></issues:blockedOn><issues:id>5</issues:id><issues:label>Type-Enhancement</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>New</issues:status></entry><entry gd:etag='W/"CUYGSX47eCl7ImA9Wx9WGEU."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/6</id><published>2010-08-11T15:23:42.000Z</published><updated>2011-01-24T15:32:08.000Z</updated><title>Worry about link text which repeats in the entry</title><content type='html'>Right now if certain link text repeats in the entry, we might just find the first one (or all of them?). This will mess up the insertion of LINK...ENDLINK, but also possibly mess up constituency results.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/6/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=6'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/6'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:id>6</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/vedrana.jankovic/</issues:uri><issues:username>vedrana.jankovic</issues:username></issues:owner><issues:stars>2</issues:stars><issues:state>open</issues:state><issues:status>Accepted</issues:status></entry><entry gd:etag='W/"A0QBRX47eCl7ImA9Wx9XEUQ."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/7</id><published>2010-11-10T20:45:54.000Z</published><updated>2011-01-05T03:49:14.000Z</updated><title>Check and fix (if necessary) insertion of <<<LINK...LINK; tags in parses in db</title><content type='html'>Right now I believe we enter the final LINK; tag on a new line after the last word of a link. If this is true, it won't distinguish between multiple words on that last line and tell us how far the link actually went.
Task: check whether this is a real problem or not and, if it is, rewrite it so that we enter LINK; right after the last word, or after the last word's node.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/7/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=7'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/7'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2011-01-05T03:49:14.000Z</issues:closedDate><issues:id>7</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/vedrana.jankovic/</issues:uri><issues:username>vedrana.jankovic</issues:username></issues:owner><issues:stars>2</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"A0QFSH47eCl7ImA9Wx9XFUk."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/8</id><published>2010-12-04T03:41:51.000Z</published><updated>2011-01-09T05:01:59.000Z</updated><title>Move to Stanford CoreNLP?</title><content type='html'>Stanford released a unified release of some of their NLP tools, including the parser.
http://nlp.stanford.edu/software/corenlp.shtml
While I don't know if we'll ever use those other tools, this clearly will be the package which will be updated and refined over time... should we move to this new package, and update our references accordingly? Or should we avoid it as, being a larger package, it may have worse memory characteristics?</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/8/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=8'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/8'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:closedDate>2011-01-09T05:01:59.000Z</issues:closedDate><issues:id>8</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>WontFix</issues:status></entry><entry gd:etag='W/"CkMASH47eCl7ImA9Wx9XEko."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/9</id><published>2011-01-05T23:34:09.000Z</published><updated>2011-01-05T23:34:09.000Z</updated><title>Figure out the hs_err_pid*.log files in the constituency locker</title><content type='html'>For some reason, when my back is turned, the constituency locker spews and collects a bajillion hs_err_pid*.log files. These seem to be the process id's of the java parser... maybe?
This task must be accepted by someone with athena access.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/9/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=9'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/9'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:id>9</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Low</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>New</issues:status></entry><entry gd:etag='W/"DUIFQH47eCl7ImA9Wx9XFUk."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/10</id><published>2011-01-09T04:31:51.000Z</published><updated>2011-01-09T04:31:51.000Z</updated><title>display.php yields weird results on some entries</title><content type='html'>http://constituency.mit.edu/display.php?start=88518&amp;end=88518 clearly displays weirdly - the tree is a line, not a tree, and everything is collapsed. Maybe phpSyntaxTree hit a size limit or something?
</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/10/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=10'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/10'/><author><name>patrick.hulin</name><uri>/u/patrick.hulin/</uri></author><issues:id>10</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>0</issues:stars><issues:state>open</issues:state><issues:status>Accepted</issues:status></entry><entry gd:etag='W/"DkYMQX47eCl7ImA9Wx9XFkU."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/11</id><published>2011-01-10T18:25:39.000Z</published><updated>2011-01-10T18:29:40.000Z</updated><title>Distinguish real "cross-clausal" failures and "multiple sentences"</title><content type='html'>Right now we don't distinguish between &quot;x_clausal&quot; failures which are *actually* links which cross over multiple clauses, but don't subsume all those clauses, and those which are simply links over multiple sentences.
Task: If the link is multiple sentences, make it pass constituency.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/11/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=11'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/11'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:id>11</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Low</issues:label><issues:label>Component-Constituency</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>2</issues:stars><issues:state>open</issues:state><issues:status>New</issues:status></entry><entry gd:etag='W/"DkYGSH47eCl7ImA9Wx9XFkU."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/12</id><published>2011-01-10T18:28:49.000Z</published><updated>2011-01-10T18:28:49.000Z</updated><title>Make punctuation pass work on exclamation points</title><content type='html'>Right now it looks like we're not stripping exclamation points correctly in our punctuation pass, resulting in some superfluous &quot;x_clausal&quot; (cross-clausal) failures. For example:
entry 1620: &quot;OUCH!!!!&quot;
That link right now results in &quot;not_constituent&quot;, &quot;x_clausal&quot;.
Task: handle these extra exclamation points in the punctuation pass.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/12/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=12'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/12'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:id>12</issues:id><issues:label>Type-Task</issues:label><issues:label>Priority-Medium</issues:label><issues:label>Component-Constituency</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>New</issues:status></entry><entry gd:etag='W/"C04CRX47eCl7ImA9Wx9bEk4."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/13</id><published>2011-01-10T18:36:10.000Z</published><updated>2011-02-20T19:59:24.000Z</updated><title>Write our own sentence-breaking code to preprocess text before giving it to the parser</title><content type='html'>Right now a number of poor parses result from the parser not accurately breaking sentences. We should try to write something ourselves to break sentences with linebreaks before giving it to the parser. This will also make the parses faster in `parse_entries.php`.
Note: in terms of the target accuracy of the sentence-breaker, it is safer to have this code *under*break sentences than to *over*break. That is, we will still get the benefit of breaking sentences (more accurate parses, faster parsing) by pre-breaking just some sentences for the parser, but if we *over*break, we could actually introduce error.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/13/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=13'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/13'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:closedDate>2011-02-20T19:59:24.000Z</issues:closedDate><issues:id>13</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-High</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"CEcMRH47eCl7ImA9Wx9WEkQ."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/14</id><published>2011-01-10T18:48:06.000Z</published><updated>2011-01-17T19:21:25.000Z</updated><title>Put better DP structure in our parses</title><content type='html'>One way to do this is the following:
1. Get the Penn Treebank 3
2. Update the Treebank using [http://sydney.edu.au/engineering/it/~dvadas1/?Noun_Phrases David Vadas's code].
3. Retrain our parser on this version of the Treebank.
4. Reparse everything.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/14/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=14'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/14'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:closedDate>2011-01-17T19:21:25.000Z</issues:closedDate><issues:id>14</issues:id><issues:label>Type-Task</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"CkANSX47eCl7ImA9Wx9XFkQ."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/15</id><published>2011-01-10T18:57:43.000Z</published><updated>2011-01-10T20:19:58.000Z</updated><title>Implement faster replacement for our tokenizer</title><content type='html'>When we want to get tokenized versions of strings, right now we simply use the Stanford LexicalizedParser class with outputFormat=words, but that means we're actually parsing everything behind the scenes. This is horribly inefficient.
Replace this code with just the tokenizer included in the stanford parser, or write our own.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/15/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=15'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/15'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:id>15</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Low</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>New</issues:status></entry><entry gd:etag='W/"CE4DQ347eCl7ImA9Wx9WEkU."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/16</id><published>2011-01-12T17:13:27.000Z</published><updated>2011-01-17T16:49:32.000Z</updated><title>judge_constituency often dies with a segfault</title><content type='html'>judge_constituency often prematurely dies, choking on particular links. This seems to be a problem with the parser.
We should first identify a set of test cases (strings which cause the parser to immediately segfault). Possible next steps include reporting this to the authors of the stanford parser.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/16/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=16'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/16'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:closedDate>2011-01-17T16:49:32.000Z</issues:closedDate><issues:id>16</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"AkAMQn47eCl7ImA9Wx9WFEs."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/17</id><published>2011-01-19T20:26:23.000Z</published><updated>2011-01-19T20:26:23.000Z</updated><title>Check intermittent failures in test suite</title><content type='html'>Some of our tests fail from time to time, non-deterministically.
Identify which tests these are, and rewrite them, or figure out why this happens.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/17/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=17'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/17'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:cc><issues:uri>/u/vedrana.jankovic/</issues:uri><issues:username>vedrana.jankovic</issues:username></issues:cc><issues:cc><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:cc><issues:id>17</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-Low</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>Accepted</issues:status></entry><entry gd:etag='W/"Ck4GRH47eCl7ImA9Wx9bEk4."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/18</id><published>2011-02-01T21:23:53.000Z</published><updated>2011-02-20T19:42:05.000Z</updated><title>Write a Condor file for interfacing with Condor compute clusters.</title><content type='html'>Running our parsing scripts on single computers is a very inefficient way to go through our database. The task is easily parallelizable, so we should be able to use a compute cluster fairly easily. MIT's CSAIL has a cluster which uses the Condor system, so we should write a script to allow us to submit jobs to it.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/18/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=18'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/18'/><author><name>patrick.hulin</name><uri>/u/patrick.hulin/</uri></author><issues:closedDate>2011-02-20T19:42:05.000Z</issues:closedDate><issues:id>18</issues:id><issues:label>Type-Enhancement</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>WontFix</issues:status></entry><entry gd:etag='W/"CUEDRX47eCl7ImA9WhZSFk0."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/19</id><published>2011-02-20T19:11:03.000Z</published><updated>2011-03-31T20:34:34.000Z</updated><title>Write flexible argument-parsing</title><content type='html'>We need more flexibility than the current argument-parsing code, which just uses argv and $_GET directly. Write code which will be easily extensible, allowing for more options. PHP does not appear to have a built-in library for this.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/19/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=19'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/19'/><author><name>patrick.hulin</name><uri>/u/patrick.hulin/</uri></author><issues:cc><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:cc><issues:closedDate>2011-03-31T20:34:34.000Z</issues:closedDate><issues:id>19</issues:id><issues:label>Type-Enhancement</issues:label><issues:label>Priority-Medium</issues:label><issues:label>Component-Constituency</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"CUECQH47eCl7ImA9WhZSFk0."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/20</id><published>2011-02-20T19:39:34.000Z</published><updated>2011-03-31T20:34:21.000Z</updated><title>Enable parallelizing</title><content type='html'>The task of running the scripts on the database is clearly parallelizable. Write a script to parallelize parse_entries and judge_constituency.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/20/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=20'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/20'/><author><name>patrick.hulin</name><uri>/u/patrick.hulin/</uri></author><issues:closedDate>2011-03-31T20:34:21.000Z</issues:closedDate><issues:id>20</issues:id><issues:label>Type-Enhancement</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>Fixed</issues:status></entry><entry gd:etag='W/"CUYFQn47eCl7ImA9WhZUFEo."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/21</id><published>2011-02-23T21:24:02.000Z</published><updated>2011-06-07T18:58:33.000Z</updated><title>Keep track of "who's touching/doing what" on central database</title><content type='html'>As we move towards having a better way to parallelize things (#20) and even run on multiple machines at once, we'll also probably want to have a way to make sure we don't step on each other's (other processes') toes.
Here are two things I think we want to keep track of on the server:
1. &quot;Task/entry ranges&quot; which are &quot;reserved&quot; by individual processes. We could highlight and visualize these on our status thing too. The script should check on startup whether it will overlap with another process, and halt if so.
2. Some data to indicate what version/setup of our code was being used on every modification to the parse and link tables... we have a timestamp now, but we will want to make sure we know exactly how to reproduce all our results. Perhaps this is just the hg revision number, perhaps not.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/21/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=21'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/21'/><author><name>mitcho%[email protected]</name><uri>/u/100307752640907358556/</uri></author><issues:cc><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:cc><issues:id>21</issues:id><issues:label>Type-Enhancement</issues:label><issues:label>Priority-Medium</issues:label><issues:owner><issues:uri>/u/100307752640907358556/</issues:uri><issues:username>mitcho%[email protected]</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>open</issues:state><issues:status>Accepted</issues:status></entry><entry gd:etag='W/"AkAERn47eCl7ImA9WhZSFk0."'><id>http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/22</id><published>2011-03-31T20:33:45.000Z</published><updated>2011-03-31T21:58:27.000Z</updated><title>Initial queries in judge_constituency run really, really slowly</title><content type='html'>The initial queries in judge_constituency run incredibly slowly, especially when running more than one of them at once.</content><link rel='replies' type='application/atom+xml' href='http://code.google.com/feeds/issues/p/hyperlink-constituency/issues/22/comments/full'/><link rel='alternate' type='text/html' href='http://code.google.com/p/hyperlink-constituency/issues/detail?id=22'/><link rel='self' type='application/atom+xml' href='https://code.google.com/feeds/issues/p/hyperlink-constituency/issues/full/22'/><author><name>patrick.hulin</name><uri>/u/patrick.hulin/</uri></author><issues:closedDate>2011-03-31T21:58:27.000Z</issues:closedDate><issues:id>22</issues:id><issues:label>Type-Defect</issues:label><issues:label>Priority-High</issues:label><issues:owner><issues:uri>/u/patrick.hulin/</issues:uri><issues:username>patrick.hulin</issues:username></issues:owner><issues:stars>1</issues:stars><issues:state>closed</issues:state><issues:status>WontFix</issues:status></entry></feed>