stmatth
Karma: 1
|
beta 8 feedback - 2010/05/27 13:08
Hi,
thanks for the beta version with all the improvements on the alias and manual modification, this is really great. I didn't have time to test it earlier, so sorry that I come in quite late.
Generally it really looks good. Here are some things I noticed:
- the plugin seems to add spaces in front and after the term. This doesn't matter in most cases but looks strange when the term is, for example, in brackets: (term) becomes ( term )
- it somehow gets confused when the term and also an alias occur in the text. For example, I have "anterior cingulate cortex" with alias "ACC", and in the text there is a "anterior cingulate cortex (ACC)" which becomes "0 (ACC)", with both the 0 and the ACC a link showing the correct definition. Somehow the first instance of "anterior cingulate cortex" gets replaced with "0".
- I haven't found documentation if it is possible to define several aliases for the same term. I've tried separating them with | (vertical bar) in the alias field but I'm not sure it works (haven't really had time to play with this.
- In the plugin, there is an option to set a regex separator. Couldn't find info what this is for - does it mean one can use regex somewhere in the term definition, or in the text?
But I think these are minor issues and Glossary has made an enormous jump forward with this beta, great work!
Cheers
Stephan
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/05/27 13:22
Hi,
further to the third point (setting multiple aliases) - I see now that it does indeed work with separating them by a vertical bar.
However there is a problem that when "find exact terms" is selected in the plugin, it doesn't seem to find terms that have an alias (but it does find terms without alias).
Hope this helps
Cheers
Stephan
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/05/30 13:10
Hi,
here is another one: When the definition contains a quote, the definition in the popup stops at that point. Reason: Quotation marks are escaped with a backslash, but should be escaped as "e;
I think this is done in glossary_plugin_content.php in the loop in lines 98-107, but my PHP is not good enough and I can't follow precisely how this replacement is happening, so I can't suggest a solution here.
Post edited by: stmatth, at: 2010/05/30 13:40
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/05/30 13:25
Hi, regarding the first item in my original mail: The plugin introduces a space before and after the term. This causes problems for example with punctuation ("... (term) ..." becomes "... ( term ) ...).
The issue comes from lines 174-186 in glossary_plugin_content.php. It is solved when you remove the empty lines and also the tabs (unfortunately this makes the code look weird as the indentation gets lost, but I think this is acceptable). So it should look like:
| Code: |
return <<<a class="glossarylink" href="$ref" title="$desc">$termasfound</a>
IMG_LINK;
}
return <<$termasfound
ONE_LINK;
|
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/05/30 13:42
in my last post, the code is all messed up. Doesn't matter, the point is simply to remove the empty lines and the tabs at the beginning of the lines so that the A-tag stands on its own, without any whitespace characters before or after it.
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
admin
Karma: 98
|
Re:beta 8 feedback - 2010/06/02 11:04
Thanks for the useful feedback - I will work through those and attempt corrections.
Martin Brampton aka Counterpoint
http://aliro.org
http://black-sheep-research.com |
|
|
| | Sorry, you do not currently have permission to write here. |
admin
Karma: 98
|
Re:beta 8 feedback - 2010/06/07 16:46
Most of these should be fixed in beta 9.
Spaces have been left around the image, but not after the term, and not before the term if images are not being used.
The appearance of 0 was a problem that occurred when more than one term appeared in close proximity. Should be fixed now.
All aliases and terms should be found and marked.
Quotes are a really difficult problem and are not fixed in beta 9. The problem is that there is no consistency about how they are handled. They can finish up in articles as quotes, as the ' element, as the &039; element, or as the &39; element. The same is probably true for some other symbols, which makes it difficult to implement any kind of standardizing approach.
In theory, I could process the article to standardize the HTML elements, but that is risky as it is impossible to be confident about parsing the HTML so as to avoid messing things up. If it was possible to be sure the article was valid XHTML, it could be analyzed as XML, but that's risky too!
Post edited by: admin, at: 2010/06/07 22:51
Martin Brampton aka Counterpoint
http://aliro.org
http://black-sheep-research.com |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/06/08 11:23
Hi Martin,
thanks. Looks good. Most of the issues have disappeared (except the " which you explain).
It finds almost all terms. There is one case where it doesn't find the term. As far as I can see, this happens when the text contains only the alias but not the term. As soon as I add the term somewhere in the text, it also finds the alias.
I think you mentioned that there is some preprocessing that filters which terms are processed, so perhaps that doesn't include the aliases?
Otherwise very good work, thank you!
Cheers
Stephan
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/06/08 11:46
Hi again,
regarding the quotation marks, isn't this character special because you are using the title in the a tag, so it is interpreted as the end of the title?
Anyway, I thought about a popup that only uses dhtml (no javascript), and came up with a simple solution. This would solve this problem and also give styling and images back.
It uses an invisible div element that is made visible at onmouseover or onclick, and it should be fairly robust.
Would be great if you could put this in as an alternative. It's attached here: http://remository.com/forum/func,view/id,16295/catid,12/
Cheers
Stephan
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
admin
Karma: 98
|
Re:beta 8 feedback - 2010/06/08 14:21
Well, I confuse myself over the quotes issue at times! But I think there is a quite tricky problem. The issue is finding terms that include quotes - there's no knowing exactly what to look for. How a quote/apostrophe will appear in the article that is passed to the Glossary content plugin depends on how the article was created. For example, different editors behave differently in whether they will turn what someone types as an apostrophe into an HTML element, and if so which one. May have to keep thinking about that one.
Thanks for your work on fancier popups - I'll try to incorporate that into Glossary when I can.
Picking up definitions for aliases in content was quite a major bug, but I think it is now fixed. I've also made significant progress on accented characters, which is a big issue for some people. With most things working, I've decided to release the latest code as Glossary 2.7 RC. I hope my optimism will be justified!
Martin Brampton aka Counterpoint
http://aliro.org
http://black-sheep-research.com |
|
|
| | Sorry, you do not currently have permission to write here. |
stmatth
Karma: 1
|
Re:beta 8 feedback - 2010/06/08 18:16
Hi, I tested RC and couldn't find any obvious problems. It finds all the terms now and everything looks fine. Well done!
Only the quote issues are still there. I see what you mean now, I didn't think of quotes in the term. However I suspect the more common problem is a quote in the definition. Anyway, the user can easily avoid this issue by using the typographic “ and ” (or „ in some languages) when writing the definitions.
It would be great if you look at including markup and images again some time, but I understand it's better if you want to concentrate on getting everything else finished now.
Keep up the good work!
Cheers
Stephan
Stephan Matthiesen - Science-Texts: editorial team for scientific texts |
|
|
| | Sorry, you do not currently have permission to write here. |
|
Forum
