Add ability to create publication lists

toomuchnoise commented 8 years ago

Currently it is possible to add publication citations by creating a reference and then bibliography at the end of a page. Often times you simply just want a curated list of publications on your pages. Example uses include a bio page listing your publications or a lab page listing publications related to your project / team / or grant.

Possible workflow:

Divide the administrative section with tabs or sections. Add one called publications (/options-general.php?page=abt-options).
Here you have the ability to add/remove publications and add arbitrary tags to them as well as manually add publications that don't exist in PubMed.
(optional) You could potentially see publications found in your defined queries here and have the ability to add individual publications to your list directly from the query.
You would then use a shortcode to insert a publication list on a given page. Something like this: [pubmed-reflist tag=myTag] or [pubmed-reflist tag=myTag2]
Potentially there could be a way to add multiple tags, and/or logic to shortcodes, but think just a simple single tag would cover most cases as you could tag publications with multiple tags.
Tags could have their own tab for management purposes.

You might also consider moving the back end link out of Settings into Tools.

If you'd like I can create a wireframe for you. I'm also willing to supply development assistance or sponsorship. My team is very experienced with the PubMed API. Let me know what would be most useful to you. You can reach me directly at matt@unifiedmediallc.com.

toomuchnoise commented 8 years ago

Looking over your last release notes I see a strong desire to not use shortcodes. The same idea above could be implemented but rather than insert the list via a shortcode you could simply place your cursor at the place of insertion and add publications through the Reference List panel with the addition of a button to insert tagged lists. Lists could still be maintained the same way as described above.

The only catch is if you update your publication list I'm not sure how the page gets the update without using a shortcode.

dsifford commented 8 years ago

Hi @toomuchnoise thanks for the feedback.

A few thoughts:

Looking over your last release notes I see a strong desire to not use shortcodes.

This is correct. The shortcode API I think was originally intended as a way for individuals to add quick functionality to their own websites, rather than for consumption in plugins or themes. I'm against the use of them in distributed plugins because it automatically creates a breaking dependency on the plugin for the user.

Another reason for not wanting to use shortcodes is because I abhor PHP to a level almost indescribable. A PHP codebase is just not something that I'd care to maintain. To this point, aside from the options page, the only thing that I've been using PHP for is serving javascript, and that's suited the project well thus far.

The same idea above could be implemented but rather than insert the list via a shortcode you could simply place your cursor at the place of insertion and add publications through the Reference List panel with the addition of a button to insert tagged lists.

Something that would likely be doable is just adding a button to the main editor that opens a panel where the user could create and manipulate a list of publications and insert them into the document after the manipulation is completed. I'd have to think about it more to keep the workflow scalable, but something along those lines might work.

The only catch is if you update your publication list I'm not sure how the page gets the update without using a shortcode.

This would work the same as how the references are updated currently. Behind the scenes, the application state uses an observable/observer pattern and all dependent children of each higher-order observable is notified of changes in state as the changes occur. This allows the entire application to react to a single change in data.

All that would need to happen would be to store the publication list state in a separate observable data store than the reference list and after changes occur to the state (from user interactions in the dialog window), the document just needs to repaint with the changes.

All that said, there's still a few things I'd like to finish up with the current feature-set prior to moving on to another feature. Namely, the dialog windows are ugly, the test coverage is atrocious, and the data flow needs optimization. I'll keep this on the radar though. You're welcome to send in a PR if you'd like, keeping in mind that I'm pretty opinionated on keeping the observable pattern.

I'll update the contributing guide later on to reflect these opinions. I'm realizing now that I don't really mention that. I'll also mention the core technologies that this application is built with, which are Typescript, React, and Mobx.

Thanks again for the feedback. Keep the great ideas comin'!

englebip commented 8 years ago

I would love to see this as well! I came across this plugin when building a site for an academic research group. It is definitely miles ahead of the other plugins I found, however the main purpose I would like it to serve is to build a list of publications from the group. This list can be procured as a RIS (from Mendeley, for example), but then:

the bibliography should be reverse sorted by date of publication (i.e., most recent first) [could be achieved in the RIS source file]
no citation numbers to be included in the main text, or (ideally) numbered in reverse, so that publication keep their number as new ones are added on top

Thanks for your work, I really appreciate it!

EDIT: typo.

EDIT2: I was wondering if the 2 issues I mentioned (bibliography sorting and citation format) couldn't be sorted by some custom CSL?

danieldekay commented 8 years ago

Would love to see something like this as well.

dsifford commented 8 years ago

Thanks for the suggestions everybody.

Just got home from vacation this week. I'll read through these more and see if I can work something out. It shouldn't be too difficult (in theory). However, the main difficulty would be in applying the feature without cramming the UI.

Will keep everybody posted as progress is made.

Of note, I still need to finish up the things mentioned here before diving into another feature.

dsifford commented 8 years ago

To the group:

Before diving into this, can you explain to me what exactly makes using American Psychological Association 6th edition (curriculum vitae, sorted by descending date) a non-viable option for satisfying your needs?

I'm suspecting the issue boils down to 2 key elements:

The plugin injects inline citations when, in fact, they aren't wanted/needed.
You'd like to have more than one single list of references.

Is that accurate?

Thanks for the clarification. 👍

dsifford commented 8 years ago

Update: Assuming my hunch above was correct, I've nearly completed a working implementation of this. Demo to follow shortly.

dsifford commented 8 years ago

@toomuchnoise @englebip @danieldekay Let me know if this satisfies what it is you all were looking for: https://vid.me/6ggS

Not going to push any updates until I hear back. Thanks 👍

toomuchnoise commented 8 years ago

This look great. A couple of questions:

Is it possible to have the same publication on more than one list on a page, or multiple lists and as a citation?
Does the interface scale to support long lists? 50 or more publications?

Matt

On Sat, Aug 27, 2016 at 5:44 PM, Derek Sifford notifications@github.com wrote:

@toomuchnoise https://github.com/toomuchnoise @englebip https://github.com/englebip @danieldekay https://github.com/danieldekay Let me know if this satisfies what it is you all were looking for: https://vid.me/6ggS

Not going to push any updates until I hear back. Thanks 👍

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-242942548, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jpwpkDk4ePUSBuQ5nCB4JajUR8Jtks5qkK-3gaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

@toomuchnoise Yes and yes.

The only caveat (perhaps) is that once you generate a static list, it persists in a detached state from the state of the reference list.

In other words, say you have a reference list that looks like this:

Cited:
-----------
ref e
ref o
===========
Uncited:
-----------
ref a
ref i
ref u

and a document that looks like this:


[...text]

static bibliograhy
1. ref a
2. ref e
3. ref o
4. ref u

[...more text]

main bibliography
1. ref e
2. ref o

Now, lets say you decide to delete ref e, ref o, and ref u from the reference list.

Here's what your reference list will look like now (Well, at least until I get to #130)

===========
Uncited:
-----------
ref a
ref i

Here's what your document will look like.


[...text]

static bibliograhy
1. ref a
2. ref e
3. ref o
4. ref u

[...more text]

So, in short, your static lists will persist exactly as it was when you first insert it until you manually delete it, no matter what the state of the reference list looks like.

Hope that makes sense!

toomuchnoise commented 8 years ago

Makes sense. Thank you. Looking forward to the commit.

Matt

On Sun, Aug 28, 2016 at 9:38 PM, Derek Sifford notifications@github.com wrote:

@toomuchnoise https://github.com/toomuchnoise Yes and yes.

The only caveat (perhaps) is that once you generate a static list, it persists in a detached state from the state of the reference list.

In other words, say you have a reference list that looks like this:

Cited:

ref e

ref o

Uncited:

ref a ref i ref u

and a document that looks like this:

[...text]

static bibliograhy

ref a

ref e

ref o

ref u

[...more text]

main bibliography

ref e

ref o

Now, lets say you decide to delete ref e, ref o, and ref u from the reference list.

Here's what your reference list will look like now (Well, at least until I get to #130 https://github.com/dsifford/academic-bloggers-toolkit/issues/130)

Uncited:

ref a ref i

Here's what your document will look like.

[...text]

static bibliograhy

ref a

ref e

ref o

ref u

[...more text]

So, in short, your static lists will persist exactly as it was when you first insert it until you manually delete it, no matter what the state of the reference list looks like.

Hope that makes sense!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-243014419, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jtW3u6wVz7rOA1sCtpoNeIdmRJL_ks5qkjgTgaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

englebip commented 8 years ago

Hi Derek,

Indeed, this would make me very happy! :+1:

And you got the issues right: no references needed in the main text, and perhaps more than one list of references is to be included.

Concerning the CSL style, yes, you got it right, something like that could work. I was thinking of using something like the ACS (American Chemical Society) format as it is relevant for the field of research of the group, but likely this can be tweaked using the style you mentioned... Will give it a shot!

Thanks for the quick reaction!

toomuchnoise commented 8 years ago

This is a fantastic update. Is there a way to edit a static list once on the page? It doesn't look like you are able to edit the static lists or change the format of the citations once inserted (though you can change citation format of the inline citations. Not sure if this was deliberate. Thank you.

On Thu, Sep 1, 2016 at 3:39 PM, Derek Sifford notifications@github.com wrote:

Closed #123 https://github.com/dsifford/academic-bloggers-toolkit/issues/123 via

128 https://github.com/dsifford/academic-bloggers-toolkit/pull/128.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#event-775669898, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jj3XYwAts505q8l8cqr0TKhC-MFBks5qlyoGgaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

@toomuchnoise Thanks!

To answer your questions....

Is there a way to edit a static list once on the page?

Nope. It's a one and done deal. If you want to add or remove references, you'll have to delete it and regenerate (any other way would be too memory-intensive, require a heap of additional meta to be saved to posts, and require a gigantic rewrite of the codebase).

[ ... ] or change the format of the citations once inserted (though you can change citation format of the inline citations. Not sure if this was deliberate.

Can you explain what you mean here? Are you saying that you'd like the citation style for the citations within the publication list to automatically update when a different style is chosen? (like the main bibliography)

If so, the fact that they don't is by design for the following reasons:

Some citation styles do not have defined bibliographies. If one of these styles is selected by mistake, you'd lose all of your publication lists.
In my mind, a larger number of users will have a need to use a different citation style for publication lists, than those who do not need to use a different style (for the reasons mentioned above, and also because some will use it exclusively for CV applications, which has its own unique style defined).
People who do not want the publication lists to have a different style than their citations should, by and large, be unaffected by this. Generally once a site picks a citation style, they stick with it. So no harm no foul.

I hope I didn't misunderstand your question. If so, let me know.

toomuchnoise commented 8 years ago

To the first point, that is unfortunate. Most people would use this feature to update a publication list which isn't static as people publish on an ongoing basis. If there are only a few in a list it's not so much of a hassle, but once the list gets long could be very cumbersome.

As I'm not familiar with the code behind the scenes for this not sure if I can offer anything constructive. Would it be possible when you click on a list to simply read back in all the PubMed IDs so you don't have to keep everything in memory all the time?

On the second point, it was more of a question rather than a request. When you create inline styles you can go back to the list at the end of the page click on it and change how the citation looks via the dropdown. This is not the case with lists. They are what you originally created them with. This really isn't an issue, but pointed it out for interface consistency and wasn't sure if it was a bug.

Again, thank you. This is a great project.

On Thu, Sep 1, 2016 at 4:45 PM, Derek Sifford notifications@github.com wrote:

@toomuchnoise https://github.com/toomuchnoise Thanks!

To answer your questions....

Is there a way to edit a static list once on the page?

Nope. It's a one and done deal. If you want to add or remove references, you'll have to delete it and regenerate (any other way would be too memory-intensive, require a heap of additional meta to be saved to posts, and require a gigantic rewrite of the codebase).

[ ... ] or change the format of the citations once inserted (though you can change citation format of the inline citations. Not sure if this was deliberate.

Can you explain what you mean here? Are you saying that you'd like the citation style for the citations within the publication list to automatically update when a different style is chosen? (like the main bibliography)

If so, the fact that they don't is by design for the following reasons:

Some citation styles do not have defined bibliographies. If one of these styles is selected by mistake, you'd lose all of your publication lists.

In my mind, a larger number of users will have a need to use a different citation style for publication lists, than those who do not need to use a different style (for the reasons mentioned above, and also because some will use it exclusively for CV applications, which has its own unique style defined).

People who do not want the publication lists to have a different style than their citations should, by and large, be unaffected by this. Generally once a site picks a citation style, they stick with it. So no harm no foul.

I hope I didn't misunderstand your question. If so, let me know.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-244206647, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jhggNiTC9CiG6BGXmI__mEEOSqIkks5qlzlggaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

To the first point, that is unfortunate. Most people would use this feature to update a publication list which isn't static as people publish on an ongoing basis. If there are only a few in a list it's not so much of a hassle, but once the list gets long could be very cumbersome.

All the citations will still remain in your reference list, so it's really just a matter of selecting and regenerating.

If that's not a feasible option, you could always chunk your publication lists by year, being that they are a static record of the past. Once a year has completed, create another one above or below it.

As I'm not familiar with the code behind the scenes for this not sure if I can offer anything constructive. Would it be possible when you click on a list to simply read back in all the PubMed IDs so you don't have to keep everything in memory all the time?

Keeping a few (or even a few thousand) PMIDs in memory isn't difficult really at all. The issue is a bit more complex.

In order for the plugin to function smoothly (and bug-free), it has keep track of and maintain three separate "states": It's own state (all references added, removed, adjusted, etc), the state of the editor (did citations shift around? were some deleted? etc), and the state of the processor (waaaay too many things to list). If one state becomes out of sync with another, things start to go haywire. Three states is manageable, however as you begin to add more separate states for it to be responsible for, memory consumption will drive upwards exponentially. Each additional publication list would add another state to be responsible for separate from the main document state, which just can't happen.

While writing this though, I may have come up with a compromise (stress may; I'm not certain it'll even work)...

I may be able to add the ability to add references into an existing publication list. Would that be helpful?

This actually would kill two birds because this same method may also be able to attach a reference into an existing inline citation (something I've been meaning to look into anyway).

On the second point, it was more of a question rather than a request. When you create inline styles you can go back to the list at the end of the page click on it and change how the citation looks via the dropdown. This is not the case with lists. They are what you originally created them with. This really isn't an issue, but pointed it out for interface consistency and wasn't sure if it was a bug.

Ah, sorry for misunderstanding. Yep, not a bug!

Keep the ideas comin' 😄

toomuchnoise commented 8 years ago

I may be able to add the ability to add references into an existing publication list. Would that be helpful?

Yes. That would be awesome! If it was able to insert it in date order (not just append) that would be ideal.

Thank you.

On Thu, Sep 1, 2016 at 7:47 PM, Derek Sifford notifications@github.com wrote:

To the first point, that is unfortunate. Most people would use this feature to update a publication list which isn't static as people publish on an ongoing basis. If there are only a few in a list it's not so much of a hassle, but once the list gets long could be very cumbersome.

All the citations will still remain in your reference list, so it's really just a matter of selecting and regenerating.

If that's not a feasible option, you could always chunk your publication lists by year, being that they are a static record of the past. Once a year has completed, create another one above or below it.

As I'm not familiar with the code behind the scenes for this not sure if I can offer anything constructive. Would it be possible when you click on a list to simply read back in all the PubMed IDs so you don't have to keep everything in memory all the time?

Keeping a few (or even a few thousand) PMIDs in memory isn't difficult really at all. The issue is a bit more complex.

In order for the plugin to function smoothly (and bug-free), it has keep track of and maintain three separate "states": It's own state (all references added, removed, adjusted, etc), the state of the editor (did citations shift around? were some deleted? etc), and the state of the processor (waaaay too many things to list). If one state becomes of sync with another, things start to go haywire. Three states is manageable, however as you begin to add more separate states for it to be responsible for, memory consumption will drive upwards exponentially. Each additional publication list would add another state to be responsible for separate from the main document state, which just can't happen.

While writing this though, I may have come up with a compromise (stress may; I'm not certain it'll even work)...

I may be able to add the ability to add references into an existing publication list. Would that be helpful?

This actually would kill two birds because this same method may also be able to attach a reference into an existing inline citation (something I've been meaning to look into anyway).

On the second point, it was more of a question rather than a request. When you create inline styles you can go back to the list at the end of the page click on it and change how the citation looks via the dropdown. This is not the case with lists. They are what you originally created them with. This really isn't an issue, but pointed it out for interface consistency and wasn't sure if it was a bug.

Ah, sorry for misunderstanding. Yep, not a bug!

Keep the ideas comin' 😄

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-244245785, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jgZW3RFT2URxDva-lCpdcqeWKP15ks5ql2P6gaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

If it was able to insert it in date order (not just append) that would be ideal.

Yep, it would be inserted according to the specifications of your chosen citation style. Assuming it's one that sorts by date, then it'll be inserted in the proper position. 👍

dsifford commented 8 years ago

@toomuchnoise Can you create an issue spelling this out when you get time so I don't forget?

toomuchnoise commented 8 years ago

Yes. I'll add a feature request. Thank you.

On Thu, Sep 1, 2016 at 7:53 PM, Derek Sifford notifications@github.com wrote:

@toomuchnoise https://github.com/toomuchnoise Can you create an issue spelling this out when you get time so I don't forget?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-244246725, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jkDjht0UOl0kxArIahRS7JW-cNIbks5ql2WAgaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

@toomuchnoise Yep, just tried my hunch real quick and both of the above things I mentioned are going to be possible.

I'll add it by the next release. 👍

dsifford commented 8 years ago

@toomuchnoise I just finished up (for the most part) adding both of the features we discussed yesterday.

Still have a little cleanup and testing to do, but I should have the update pushed out within the next 3 hours. Going away for labor day weekend, so I'm trying to get this finished before then. 👍

Here's an example...

gif

Hope this satisfies what you were looking for.

dsifford commented 8 years ago

@toomuchnoise Out of time for now, but it's mostly working.

I know for a fact there's still a small bug somewhere because every once in a while I'm able to make things screw up. The bug isn't really a big deal though, so I'll just push out what I have now and fix whatever it is on Sunday.

That said, if you experience a bug and can figure out how to reproduce it, please please report back with the steps!

Also thanks so much for the kind donation :)

Updating the changelog and pushing out in the next 10 mins. 👍

toomuchnoise commented 8 years ago

Great. I'll take a look to see if I can trigger it.

Matt

On Friday, September 2, 2016, Derek Sifford notifications@github.com wrote:

@toomuchnoise https://github.com/toomuchnoise Out of time for now, but it's mostly working.

I know for a fact there's still a small bug somewhere because every once in a while I'm able to make things screw up. The bug isn't really a big deal though, so I'll just push out what I have now and fix whatever it is on Sunday.

That said, if you experience a bug and can figure out how to reproduce it, please please report back with the steps!

Also thanks so much for the kind donation :)

Updating the changelog and pushing out in the next 10 mins. 👍

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-244490660, or mute the thread https://github.com/notifications/unsubscribe-auth/AL9_jp80h8d4X4QHpVwjqvrnpe8aH2o9ks5qmJO-gaJpZM4JoGEr .

Matthew LaPointe, MBA web + graphics + communications matt@unifiedmediallc.com phone 703-679-7131

fax 703-649-6290

www.unifiedmediallc.com

dsifford commented 8 years ago

@toomuchnoise Back home. Any success in triggering that bug?

GAtherton commented 8 years ago

I am currently using this new function to creat static lists from RIS files without inserting into the content text (as there isn't any in this instance, we just want a list of publications for each year) - this is particularly great as it was something I was trying to do a few months back.

I have achieved great lists with style Harvard but when the list is created there are no spaces between references and no bulleting/numbering so it is difficult to read the solid mass of text. Also no way to use different fonts to create the list? If you could creat those additional feature it would be perfect!

many thanks for your great plugin - saving many hours of my time now and in the future

dsifford commented 8 years ago

@GAtherton

Hi there. Can you paste a screenshot of what you're describing?

Also, I was notified that your site triggered (at least) 3 errors when importing RIS files. Are you still having issues there or have you figured that out?

GAtherton commented 8 years ago

Certainly - screenshot attached. Indenting of the second & subsequent lines would be a good alternative to bullets & spaces. It does look better now I have viewed the lists after saving as of course styling has been applied (removing the need to comply with one of my requests - see second screenshot). Some spacing or bullets would still be a good idea.

I ran into an issue with Google Chrome - the latest version of ABT does not run properly on the latest version of Chrome. I switched to Firefox and all was well.

Having said that I have just tried to add a couple of extra references to the static list by importing via a RIS file. This went well until I clicked on 'Insert static publication list' button whereupon I got an error and the list did not update - see third screenshot

Thanks for your help

Grahamn

[image: photo] Graham Atherton National Aspergillosis Centre 0161 291 5866 | graham.atherton@manchester.ac.uk | http://www.nationalaspergillosiscentre.org.uk | http://www.aspergillus.org.uk | ERC, UHSM, Wythenshawe, Manchester M23 9LT | http://www.fungalresearchtrust.org/ | http://nacpatients.org.uk/ | http://www.aspergillosis.org/ | http://www.aspergilloma.co.uk/ http://www.facebook.com/DrGrahamAtherton http://www.linkedin.com/in/grahamatherton?trk=hp-identity-name http://twitter.com/AspergillusWeb Get a signature like this: Click here! http://ws-promos.appspot.com/r?rdata=eyJydXJsIjogImh0dHA6Ly93d3cud2lzZXN0YW1wLmNvbS9lbWFpbC1pbnN0YWxsP3dzX25jaWQ9NjcyMjk0MDA4JnV0bV9zb3VyY2U9ZXh0ZW5zaW9uJnV0bV9tZWRpdW09ZW1haWwmdXRtX2NhbXBhaWduPXByb21vXzU3MzI1Njg1NDg3Njk3OTIiLCAiZSI6ICI1NzMyNTY4NTQ4NzY5NzkyIn0=&u=379834503045490

On Thu, Sep 22, 2016 at 3:56 PM, Derek Sifford notifications@github.com wrote:

@GAtherton https://github.com/GAtherton

Hi there. Can you paste a screenshot of what you're describing?

Also, I was notified that your site triggered (at least) 3 errors when importing RIS files. Are you still having issues there or have you figured that out?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-248928468, or mute the thread https://github.com/notifications/unsubscribe-auth/AJCjxUoNJ5xFA5b9y7tSe3INIVBZ3gmOks5qspcxgaJpZM4JoGEr .

[image: photo] Graham Atherton National Aspergillosis Centre 0161 291 5866 | graham.atherton@manchester.ac.uk | http://www.nationalaspergillosiscentre.org.uk | http://www.aspergillus.org.uk | ERC, UHSM, Wythenshawe, Manchester M23 9LT | http://www.fungalresearchtrust.org/ | http://nacpatients.org.uk/ | http://www.aspergillosis.org/ | http://www.aspergilloma.co.uk/ http://www.facebook.com/DrGrahamAtherton http://www.linkedin.com/in/grahamatherton?trk=hp-identity-name http://twitter.com/AspergillusWeb Get a signature like this: Click here! http://ws-promos.appspot.com/r?rdata=eyJydXJsIjogImh0dHA6Ly93d3cud2lzZXN0YW1wLmNvbS9lbWFpbC1pbnN0YWxsP3dzX25jaWQ9NjcyMjk0MDA4JnV0bV9zb3VyY2U9ZXh0ZW5zaW9uJnV0bV9tZWRpdW09ZW1haWwmdXRtX2NhbXBhaWduPXByb21vXzU3MzI1Njg1NDg3Njk3OTIiLCAiZSI6ICI1NzMyNTY4NTQ4NzY5NzkyIn0=&u=484067975263258

dsifford commented 8 years ago

@GAtherton FYI: Email attachments don't show up in the issues thread. Can you paste it over directly?

I ran into an issue with Google Chrome - the latest version of ABT does not run properly on the latest version of Chrome.

I don't think it's chrome that's causing the issue. I'm running Chrome Canary (Version 55; the most bleeding-edge version of chrome) and I haven't been able to trigger the errors that you are experiencing.

Can you walk me through the steps that you take to trigger these errors?

GAtherton commented 8 years ago

Sure - I have created a static publication list for 2015 (see attachments shortly).

I now want to add a couple of new publications to that list that I found in Endnote/Pubmed. I export those publication references to RIS and successfully import them into ABT - I can now select them as citations in the plugin list.

I click on the 2015 list which highlights, then click to insert exactly like this

Adding references to existing static publication lists

Click an existing publication list to select to.
Select all additional references that you'd like to the selected publication list from the reference list.
Click the "hamburger menu" button [1e] https://github.com/dsifford/academic-bloggers-toolkit/wiki#visual-aids to open the menu.
Click the "Insert Static Publication List" icon to insert the selected references into the selected static publication list. [2d] https://github.com/dsifford/academic-bloggers-toolkit/wiki#visual-aids

When I click button 2d I get a dialog containing an error

Posting screenshots now

Graham

abtscreenshoterror abtscreenshot2

GAtherton commented 8 years ago

When you have a moment I found another bug - the links in the static publication lists (mostly to pubmed) all have a trailing > character which breaks them - they should not be part of the link see http://www.asperdev.co.uk/FIT/achievements for my lists

http://www.ncbi.nlm.nih.gov/pubmed/26546263> does not work http://www.ncbi.nlm.nih.gov/pubmed/26546263 does work

dsifford commented 8 years ago

@GAtherton Great, I'll wait and have a look at those screenshots when you get them uploaded.

When you have a moment I found another bug - the links in the static publication lists (mostly to pubmed) all have a trailing > character which breaks them

This was fixed a few versions ago. I can't reproduce this anymore. Can you clear your cache and see if that still happens?

dsifford commented 8 years ago

Also, what citation style are you using?

dsifford commented 8 years ago

@GAtherton

I was able to reproduce the error that triggered when additional references were added to the static pub lists. I think I have that all sorted out now.

Re: formatting of the publication lists -- I think this comes down to two factors...

Does your theme have wpautop disabled? (That's the function that automatically inserts <p> tags in your editor when you hit enter.
What citation style are you using? Can you try with american-medical-association and see if it still happens?

Re: insertion of > character in URLs -- I still cannot reproduce that.

GAtherton commented 8 years ago

Sorry at home now. I am using Harvard I think the one top of the list offered but will confirm tomorrow.

Many thanks for your work

Graham

On 22 Sep 2016 5:27 p.m., "Derek Sifford" notifications@github.com wrote:

Also, what citation style are you using?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-248955123, or mute the thread https://github.com/notifications/unsubscribe-auth/AJCjxYAFmljlvPwQ8apghpABhwSYm6iqks5qsqxTgaJpZM4JoGEr .

dsifford commented 8 years ago

@GAtherton

There are 73 variants of the Harvard style. The one closest to what you are describing is Harvard - Anglia Ruskin University which, incidentally, is also the only version that seems to be affected by the html closing bracket bug (see below screenshot).

My recommendation would be to choose from any of the other 72 options, rather than using that style since the CSL rules defining that individual style are ostensibly not correct.

GAtherton commented 8 years ago

What are the chances - well they were 73:1

Many thanks for your extremely prompt support - great plugin!

Graham

On 22 Sep 2016 8:41 p.m., "Derek Sifford" notifications@github.com wrote:

@GAtherton https://github.com/GAtherton

There are 73 variants of the Harvard style. The one closest to what you are describing is Harvard - Anglia Ruskin University which, incidentally, is also the only version that seems to be affected by the html closing bracket bug (see below screenshot).

[image: image] https://cloud.githubusercontent.com/assets/5240018/18763330/dfab8768-80da-11e6-9aec-08998361d1bc.png

My recommendation would be to choose from any of the other 72 options, rather than using that style since the CSL rules defining that individual style is ostensibly not correct.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/dsifford/academic-bloggers-toolkit/issues/123#issuecomment-249007027, or mute the thread https://github.com/notifications/unsubscribe-auth/AJCjxd3_pLkJVX2Q-SrGTQDEZiFmNzNMks5qstnTgaJpZM4JoGEr .

dsifford commented 8 years ago

@GAtherton

Thanks for your patience. Glad to hear you like the plugin!

I've fixed the closing bracket issue and will include that in the next update 👍

dsifford / academic-bloggers-toolkit