Provides an extended bookmark export #1470
Conversation
THis provides a richer export, incompatible with old browsers, but allowing nicer integration mechanisms into other systems. I typically gonna use that to generate a lifestream including my Shaarli content.
tpl/default/export.html
Outdated
| <span class="label-name">{'Extend import to include thumbnail and permalink|t}</span><br> | ||
| <span class="label-desc">{'Provide a richer export, unfortunatly incompatible with browsers'|t}</span> |
There was a problem hiding this comment.
The French translation is officially supported, so this PR should include the updated translation. I assume you're French, but let me know if you need help.
There was a problem hiding this comment.
Also, why is it incompatible with browsers? Isn't only included the description?
Does it work as export -> import into Shaarli?
There was a problem hiding this comment.
It' may incompatible with browsers in the sense that the export format reverse engineered by @sebsauvage doesn't include any supplmenetary HTML code. By adding code here, I take the risk of not maintaining that compatibility.
And frankly, I have absolutly not the courage to test it on all browsers, so this message is more a caution message than anything else.
There was a problem hiding this comment.
The French translation is officially supported, so this PR should include the updated translation. I assume you're French, but let me know if you need help.
Yup, I would gladly get some help ... do you have any tutorial on PHP gettext usage for Windows users ?
There was a problem hiding this comment.
any tutorial on PHP gettext usage for Windows users ?
Please have a look at https://shaarli.readthedocs.io/en/doc-rework-setup/dev/Translations/ and tell us if any information is missing (part of the documentation refactoring which is not merged yet)
There was a problem hiding this comment.
test it on all browsers
We should probably test it in Firefox and Chrome/Chromium on Linux/Windows so that we can make the warning more accurate. I can take care of it (it will be faster if you can send an example export .html directly here).
Does it work as export -> import into Shaarli?
Important part. Maybe add some test cases to https://github.com/shaarli/Shaarli/blob/master/tests/netscape/BookmarkImportTest.php ?
There was a problem hiding this comment.
Yes, it works exactly like the standard export feature of Shaarli.
According to ArthurHoaro, which knows better PHP than I do, this syntax is cooler ... and it is ! Co-authored-by: ArthurHoaro <[email protected]>
… wherever it will be stored.
Forgot one quote Co-authored-by: ArthurHoaro <[email protected]>
A small wording error Co-authored-by: ArthurHoaro <[email protected]>
OK, so, to test, I create the Docker image, run it to create a link, export that link, delete the data dir, import that link. Is this test procedure correct to you ? Besides, while testing that, I had to change the |
|
OK, I've tested and fixed the export template. |
This change allows to have a <a> tag in the bookmark description. See: shaarli/Shaarli#1470
|
I've run some tests, it looks fine. I ran into a few issues that were actually due to the netscape bookmark parser dependency. I have merged a fix in this project a released a new version. Can you run I tried to import it in Firefox, it's working but it also import all permalinks as distinct entries. |
|
To avoid the permalinks being imported, you should change the Netscape
importer to only read the <DD> tag direct content, and not the contained
tag, because I use a <A> tag to add permaling/icon.
And the composer.lock has been commited.
Le 06/06/2020 à 18:28, ArthurHoaro a écrit :
…
I've run some tests, it looks fine. I ran into a few issues that were
actually due to the netscape bookmark parser dependency. I have merged
a fix in this project a released a new version. Can you run |composer
update| and commit the |composer.lock| in your PR?
I tried to import it in Firefox, it's working but it also import all
permalinks as distinct entries.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#1470 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AABJYCS3IAZWYUZWVCJVT53RVJVEJANCNFSM4NFJJUHQ>.
|
|
This needs to be rebased on master |
|
Since this patch is old, and other "rich" export solutions are being discussed in #1450 (probably not using the Netscape HTML format), I suggest closing this and starting over once we know exactly what's expected (a full JSON export, without the need for a third-party API client, would do the trick in my opinion). What do you think? |
|
Closing in favor of #1450 |
THis provides a richer export, incompatible with old browsers, but allowing nicer integration mechanisms into other systems.
I typically gonna use that to generate a lifestream including my Shaarli content.
Fixes #1450