Ecoinformatics Redmine: Ryan McFallhttps://projects.ecoinformatics.org/ecoinfo/https://projects.ecoinformatics.org/ecoinfo/ecoinfo/favicon.ico?14691340362010-03-31T09:34:31ZEcoinformatics Redmine
Redmine FIRST - Bug #4917: Parsed multi-part questions aren't saved/retrieved correctlyhttps://projects.ecoinformatics.org/ecoinfo/issues/4917#change-169792010-03-31T09:34:31ZRyan McFallmcfall@hope.edu
<p>Do you have specific documents for this bug - the original PDF, the parsed XML documents representing the questions? Do you know if the user did any editing within the parser, or was the default parsed output accepted?</p> FIRST - Bug #4643: Images missing from a re-opened assessment when the images are the choices of ...https://projects.ecoinformatics.org/ecoinfo/issues/4643#change-159352010-01-14T19:12:27ZRyan McFallmcfall@hope.edu
<p>(In reply to comment <a class="issue tracker-1 status-5 priority-5 priority-highest closed" title="Bug: mde won't load because of hardcoded image paths (Closed)" href="https://projects.ecoinformatics.org/ecoinfo/issues/5">#5</a>)</p>
<blockquote>
<p>during the save, only images returned in by ResponseItem.getImageList() are<br />being saved with ecogrid URIs. It sounds like there are other images that need<br />this same treatment.</p>
<p>For multiple choice items, these images (from the choices) are not being<br />processed:<br />MultipleChoiceResponseItem.choices.get(index).getImageList()</p>
<p>I think we need to process any items that implement TextAndImage interface.<br />Seems that only MultipleChoiceResponseItem needs special processing here.</p>
</blockquote>
<p>Sandeep:</p>
<p>Can you look into this and make a proposal for how to handle? If you don't know for sure what's going on, I can assist if necessary.</p> FIRST - Bug #4643: Images missing from a re-opened assessment when the images are the choices of ...https://projects.ecoinformatics.org/ecoinfo/issues/4643#change-159302010-01-06T14:55:48ZRyan McFallmcfall@hope.edu
<p>I haven't had a chance to take a look at this problem since we discussed it a couple of weeks ago.</p>
<p>As I was thinking about it this morning, however, it seems that this is the problem that we fixed when I edited all of Jay's parsed assessments. So I'm a bit worried about that, because if I recall correctly, the problem was a somewhat strange interpretation of the appropriate XML path for the images. I was fairly confident that we had modified the code and fixed that bug, so I'm wondering why it has cropped up again.</p>
<p>Sandeep, do you have time to take a look at this and determine if an incorrect path is being specified? I'm sorry that I don't remember exactly where this would be, although I think it might be MultipleChoiceResponseItemComponent or something like that (I don't have the code accessible to me at the moment).</p> FIRST - Bug #4435: The "Assign" button is de-activated if both response and score are seleceted ...https://projects.ecoinformatics.org/ecoinfo/issues/4435#change-151552009-11-24T18:18:35ZRyan McFallmcfall@hope.edu
<p>This was fixed by selecting all data types in the (invisible) data type selection box when "one file per student" or "one file per question" is the selected option.</p> FIRST - Bug #4110: Next button not enabled correctly when going back to IdentifyStudentColumnhttps://projects.ecoinformatics.org/ecoinfo/issues/4110#change-139222009-09-30T18:07:02ZRyan McFallmcfall@hope.edu
<p>Sandeep, if you've got time to try to figure this out, it wouldn't hurt to try to figure this out. I suspect that looking at the "isNextEnabledInitially (name is approximate)" method of the class IdentifyStudentColumn would allow you to fix this one.</p> FIRST - Bug #4375: item metadata not serializing correctly (Metadata items not faithfully retriev...https://projects.ecoinformatics.org/ecoinfo/issues/4375#change-149612009-09-26T09:35:04ZRyan McFallmcfall@hope.edu
<p>This should be fixed with revision 1387; the outer loop index of a nested loop was being used to retrieve the concept name and value, so only the first item from each vocabularly was being added.</p> FIRST - Bug #4352: Metadata for each item disappears after saving and re-openinghttps://projects.ecoinformatics.org/ecoinfo/issues/4352#change-148582009-09-08T20:44:59ZRyan McFallmcfall@hope.edu
<p>(In reply to comment <a class="issue tracker-1 status-5 priority-5 priority-highest closed" title="Bug: mde won't load because of hardcoded image paths (Closed)" href="https://projects.ecoinformatics.org/ecoinfo/issues/5">#5</a>)</p>
<blockquote>
<p>Ryan, as far as the qti: namespace prefix is concerned...I'm not sure why it is<br />required. I've done some sample document validation with these fields and the<br />parsers definitely want there to be the prefixes. The only thing that I can see<br />that is different about those tags is that they are not complex types (just<br />xs:string) in the schema. I think you have to include them at this point in the<br />game</p>
</blockquote>
<p>OK, I'll go ahead and assume that this needs to be both written and included in the XPath to read in the meta data. Sandeep, if you can do a bit more testing, I think we can call this bug fixed.</p> FIRST - Bug #4352: Metadata for each item disappears after saving and re-openinghttps://projects.ecoinformatics.org/ecoinfo/issues/4352#change-148562009-09-05T11:36:01ZRyan McFallmcfall@hope.edu
<p>See my commit message for the modifications that I made. I believe this has solved the problem. Sandeep, I'd like for you to do some more testing if you don't mind; I only tested with a single case, which had one Bloom item attached to it, but it does work now.</p>
<p>The code that generates category metadata appends a qti: namespace prefix. None of the other items in the generated XML have a namespace prefix.</p>
<p>Ben, I'm not that good with XML namespaces, but I suspect that because the root element of a question file has a qti prefix (qti:questestinterop), all of its descendants will inherit its namespace and the qti: in front of the fieldlabel element is unnecessary. What does putting the qti: in front of it actually do? Is it making the name of the element actually include a qti:?</p> FIRST - Bug #4352: Metadata for each item disappears after saving and re-openinghttps://projects.ecoinformatics.org/ecoinfo/issues/4352#change-148542009-09-01T13:01:05ZRyan McFallmcfall@hope.edu
<p>Looking at the code for parseDOMNodeAsResponseItem in MultipleChoiceResponseItem, I don't see any code looking for the meta data.</p>
<p>I do see a method called parseDOMNodeAsMetadata in the abstract ResponseItem class. So I think this will be as simple as actually making a call to that method in the various ResponseItem subclasses.</p>
<p>Sandeep, can you give this a look?</p> FIRST - Bug #4146 (New): Allow for multiple responses (check all that apply) for MultipleChoiceRe...https://projects.ecoinformatics.org/ecoinfo/issues/41462009-06-11T15:26:05ZRyan McFallmcfall@hope.edu
<p>In MultipleChoiceResponseItem.generateXML, there is a possibility of allowing multiple responses. However, the code below doesn't make a lot of sense:</p>
<p>if(multipleResponses){<br /> xml.append("<response_lid ident = \"Q"+id+"\" rcardinality = \"Multiple\" rtiming = \"No\">\n");<br /> xml.append("<render_choice shuffle = \"Yes\" minnumber = \"1\" maxnumber = \"4\">\n");<br />}</p>
<p>In particular, the maxnumber attribute is just an arbitrary number, and so there needs to be a way for the user to specify this.</p> FIRST - Bug #4110 (Resolved): Next button not enabled correctly when going back to IdentifyStuden...https://projects.ecoinformatics.org/ecoinfo/issues/41102009-05-27T19:11:59ZRyan McFallmcfall@hope.edu
<p>If you set the identifier column, and then click next, the correct column will be shown, but next will not be enabled.</p> FIRST - Bug #4005: Question separator function logic fails while parsing Exam2.pdfhttps://projects.ecoinformatics.org/ecoinfo/issues/4005#change-135222009-05-12T14:42:53ZRyan McFallmcfall@hope.edu
<p>I'm not sure why you are saying that we need to know the index of the first valid line of text on the current page. We would know where we <del>think</del> the answer key starts (although I don't remember if StringBlocks contain geometry information). We could then mark up the PDF with a suggestion for the answer key by selecting the entire portion of the document from there on if we wanted to.</p>
<p>I don't think the answer key problem can be solved by getting rid of lines that look like the ones that separateQuestions uses to guess about being in the answer key. The subsequent lines after the beginning of the answer key need to be ignored as well.</p>
<p>I'd still like to see some more intelligent guessing about whether we're in the answer key or not. The presence of the words Answer Key is obviously a pretty good indication. However, the <number>) <letter> thing seems fairly arbitrary to me. It works for multiple choice questions, but not any other type. To me, it seems likely that we are in the answer key if the following conditions hold:<br /> - The out of order number that we think is the beginning of the answer key is the same number as the number for the first question<br /> - We see numbers after the first out of order number for all questions on the assessment<br /> - There aren't any numbers larger than the last one encountered before the answer key started after the answer key</p>
<pre><code>What do you think?</code></pre> FIRST - Bug #4058 (New): Question 46 on Fall05-Exam3.pdf causes error in recognitionhttps://projects.ecoinformatics.org/ecoinfo/issues/40582009-05-11T14:35:45ZRyan McFallmcfall@hope.edu
<p>(This is a resubmission of a bug lost due to fire. The old bug number was 4057; that number seems to have taken by a newer bug submitted after the restore. We talked about this bug during the videoconference on 2009-05-06).</p>
<p>Question 46 on the exam private/samples/Fall05-Exam3.pdf is not recognized<br />correctly. This question contains some text, followed by a line of text on<br />it's own line that represents a string of DNA, and then contains several<br />choices which are all numbers.</p>
<p>Not all of the prompt is put into the question text, and the last choice (E)<br />also contains the question text for the next question.</p>
<p>Sandeep, can you investigate this and make a report back as to why this is<br />occurring?</p> FIRST - Bug #4005: Question separator function logic fails while parsing Exam2.pdfhttps://projects.ecoinformatics.org/ecoinfo/issues/4005#change-135202009-05-11T14:32:04ZRyan McFallmcfall@hope.edu
<p>(Resubmit after fire)</p>
<p>In trying to handle the answer key, I see you're showing an error message in<br />PDFAssessment2_4.java and then calling System.exit () inside of the<br />separateQuestions method.</p>
<p>We definitely don't want to call System.exit; this will cause the whole Morpho<br />application to exit, and any metadata that has been entered may be deleted.</p>
<p>I think it would be better for us to define an Exception class that represents<br />a recognition exception. Then the code that is using PDFAssessment2_4 could<br />provide the user with a way to fix the error rather than making them start over<br />from scratch. This would involve showing the wizard page that specifies<br />regions to be ignored outside the context of the wizard. The class that does<br />this is SelectIndividualText in the package edu.msu.first.parser.gui.wizard.<br />It will require that you populate a WizardData object with the appropriate<br />information (it would be the information saved in AssessmentFileSelection, also<br />in that package).</p>
<p>Can you give this a shot? I can help if you need more information.</p> FIRST - Bug #4024: Image drop target disappears after deleting last image from a response itemhttps://projects.ecoinformatics.org/ecoinfo/issues/4024#change-136092009-05-11T14:16:53ZRyan McFallmcfall@hope.edu
<p>This is no longer an issue with the new image drag-and-drop target notification<br />mechanisms. (Resubmit after fire)</p>