Daily Log for #alfresco IRC Channel

Alfresco discussion and collaboration. Stick around a few hours after asking a question.

Official support for Enterprise subscribers: support.alfresco.com.

Joining the Channel:

Join in the conversation by getting an IRC client and connecting to #alfresco at Freenode. Our you can use the IRC web chat.

More information about the channel is in the wiki.

Getting Help

More help is available in this list of resources.

Daily Log for #alfresco

2018-02-05 08:02:37 GMT <twen> hello

2018-02-05 08:58:37 GMT <yreg> Morning everyone

2018-02-05 10:15:10 GMT <qwebirc10599> I need to customise the translation in the file share/WEB-INF/classes/alfresco/site-webscripts/org/alfresco/share/imports/share-header.lib_de.properties

2018-02-05 10:15:55 GMT <qwebirc10599> this file is imported by Aikau

2018-02-05 10:16:03 GMT <qwebirc10599> what is the best way to customise it ?

2018-02-05 10:23:20 GMT <yreg> qwebirc10599, $TOMCAT_HOME/shared/classes/alfresco/site-webscripts/org/alfresco/share/imports/share-header.lib_de.properties

2018-02-05 10:23:28 GMT <yreg> should be fine

2018-02-05 10:25:33 GMT <qwebirc10599> yreg, I do not think that it works because these files are imported

2018-02-05 10:25:58 GMT <qwebirc10599> and cannot be overwritten like all the others

2018-02-05 10:26:29 GMT <qwebirc10599> they are like any other .lib files

2018-02-05 10:26:33 GMT <Tichodroma> The OOXML transformer tries to extract an embedded thumbnail from a docx which fails if there is none. Id' like to disable this shortcut and force the transformer to generate the doclib thumbnail from the first page of the pdf thumbnail.

2018-02-05 10:26:51 GMT <qwebirc10599> for these you usually need to customise the file that is importing them

2018-02-05 10:26:57 GMT <Tichodroma> this approach of AFaust does not have any effect: https://community.alfresco.com/thread/233933-preview-of-wordexcelpowerpoint-files-doesnt-work#comment-822928

2018-02-05 10:26:59 GMT <alfbot> Title: Preview of Word/Excel/PowerPoint files doesn't ... | Alfresco Community (at community.alfresco.com)

2018-02-05 10:27:05 GMT <Tichodroma> Ideas how to achieve this?

2018-02-05 10:28:26 GMT <AFaust> Tichodroma: What does TransformerDebug say about the transformers tried and the order they are tried in?

2018-02-05 10:28:33 GMT <Tichodroma> let me check

2018-02-05 10:29:10 GMT <Tichodroma> hm, the debug level was not low enough for this

2018-02-05 10:29:15 GMT <AFaust> Because what I suggested in that thread will only work as long as there are other transformers that do not "decline" to execute...

2018-02-05 10:29:29 GMT <Tichodroma> sounds reasonable

2018-02-05 10:29:38 GMT <AFaust> No need for debug level if you use the ootbee-support-tools addon and use the admin console tool for transformer tests

2018-02-05 10:29:45 GMT <Tichodroma> :)

2018-02-05 10:31:35 GMT <AFaust> The only thing that tool can't / doesn't to at this time is test with a specific node provided via a reference - it only uses pre-packaged test documents

2018-02-05 10:33:19 GMT <Tichodroma> AFaust: only TikaAuto

2018-02-05 10:35:32 GMT <AFaust> For comparison sake: Which Alfresco version and which specific file extensions are used? I assume docx => jpg?

2018-02-05 10:35:50 GMT <Tichodroma> 5.2f, docx, png (doclib)

2018-02-05 10:36:10 GMT <Tichodroma> see https://screenshots.firefox.com/hVX1525AdNcOwz1s/alf-krzn-511.ecm4u.intra

2018-02-05 10:36:11 GMT <alfbot> Title: Screenshot: ⁨Alfresco » Transformationen [Host: alf-krzn-511.ecm4u.intra IP:]⁩ (at screenshots.firefox.com)

2018-02-05 10:36:57 GMT <Tichodroma> same for docx => jpg: https://screenshots.firefox.com/x3BOWNejxW5JbyCy/alf-krzn-511.ecm4u.intra

2018-02-05 10:36:58 GMT <alfbot> Title: Screenshot: ⁨Alfresco » Transformationen [Host: alf-krzn-511.ecm4u.intra IP:]⁩ (at screenshots.firefox.com)

2018-02-05 10:37:02 GMT <fwu> hi all!

2018-02-05 10:38:44 GMT <AFaust> Starting my 5.2 instance now. Just checked in my 5.0 and saw complex.OpenOffice.Image being attempted before OOXMLThumbnail

2018-02-05 10:40:49 GMT <AFaust> Ok, 5.2 produces same result for me like shown in your screenshot.

2018-02-05 10:41:19 GMT <Tichodroma> good to know :)

2018-02-05 10:42:29 GMT <AFaust> But looking at default transformers.properties it is not immediately clear to me why...

2018-02-05 10:42:36 GMT <Tichodroma> true

2018-02-05 10:43:21 GMT <AFaust> Based on "Transformation by extension" the complex.OpenOffice.Image is still eligible...

2018-02-05 10:44:21 GMT <AFaust> Wait.... second run on the test transform now works and provides different result... Looks like not all transformers had been properly initialised after my quick server start

2018-02-05 10:44:35 GMT <Tichodroma> !

2018-02-05 10:45:26 GMT <AFaust> https://snag.gy/fHmB25.jpg

2018-02-05 10:46:15 GMT <AFaust> It now properly uses complex.OpenOffice.Image with pipeline of OpenOffice and alfresco-pdf-renderer

2018-02-05 10:47:09 GMT <Tichodroma> AFaust: do you have the alfresco-pdf-renderer binary in your installation? My 5.2 is missing it on startup.

2018-02-05 10:47:41 GMT <AFaust> I have it in my installation because I installed it... Keep in mind I never use any of the installers Alfresco provides...

2018-02-05 10:47:59 GMT <Tichodroma> of course :)

2018-02-05 10:51:19 GMT * AFaust is weirded out by the incorrect IP reported in admin console...

2018-02-05 10:52:58 GMT <AFaust> Hrmpf... great. Appears Alfresco code simply picks any interface and used the DockerNAT interface IP by chance / virtue of ordering...

2018-02-05 10:57:28 GMT <Tichodroma> adding alfresco-pdf-renderer to the path solves the problem

2018-02-05 11:09:01 GMT <fwu> ppl, anyone using Alfresco in a case management context?

2018-02-05 11:09:45 GMT <Tichodroma> Yes, but we don't store case information in Alfresco. We link to cases in an extarnal system which is the *leading* system for cases.

2018-02-05 11:10:49 GMT <fwu> Tichodroma, but that external system is a custom system?

2018-02-05 11:11:30 GMT <Tichodroma> It is a system that provides an API that we can connect to.

2018-02-05 11:11:38 GMT <Tichodroma> It could be EspoCRM, for example.

2018-02-05 11:11:45 GMT <hi-ko> fwu: we use an abstraction layer to access any case system

2018-02-05 11:12:37 GMT <hi-ko> and this layer (master data hub) has a connector to e.g. EspoCRM

2018-02-05 11:14:33 GMT <hi-ko> Alfresco uses several extensions to make the integration as smooth as possible: JS-API to read and write over the master data hub, Share controls and Share Folder controls to embed external data as if stored in Alfresco

2018-02-05 11:15:35 GMT <Tichodroma> fwu: What is your use case?

2018-02-05 11:19:47 GMT <fwu> I understand a "case" as a superset of documents + workflows + collaboration ... all in a main folder. Something like a share site.

2018-02-05 11:20:35 GMT <fwu> but a site doesnt seems a good option. I read a lot of people complaining about it as soon as they have many sites.

2018-02-05 11:21:20 GMT <fwu> so, I was thinking about something more simple: like a custom relation between workflows and documents.

2018-02-05 11:24:10 GMT <Tichodroma> The problem I see with this approach is that you can only store metadata on nodes. Which node would be used for this?

2018-02-05 11:24:40 GMT <Tichodroma> That's why we move metadata about the case to an external system like EspoCRM and only reference a case in this system on an Alfresco node.

2018-02-05 11:25:06 GMT <Tichodroma> But information like name, addresse, contact information for the case is never stored in Alfresco

2018-02-05 11:34:25 GMT <fwu> Tichodroma: what about a simple extension in a custom table, or something like that? I would like to avoid another system due to effort integration and UI related inconsistencies between two solutions.

2018-02-05 11:35:18 GMT <fwu> I think Alvex is doing something like this with the idea of "related workflows"

2018-02-05 11:36:51 GMT <Tichodroma> Do you mean this? http://www.alvexcore.com/

2018-02-05 11:36:52 GMT <alfbot> Title: Alvex (at www.alvexcore.com)

2018-02-05 11:38:02 GMT <fwu> Tichodroma, answering to your question: I would keep the main case data in the properties of the main case folder.

2018-02-05 11:38:28 GMT <fwu> yes, they implement the concept of related workflows. I believe they are using a "simple" custom table for that.

2018-02-05 11:38:45 GMT <fwu> of course this are only ideas...

2018-02-05 11:39:12 GMT <fwu> that is why Im talking with you guys!

2018-02-05 11:39:18 GMT <fwu> these

2018-02-05 11:39:21 GMT <Tichodroma> Alvex looks like it heavily depends on Share which has no future.

2018-02-05 11:40:01 GMT <Tichodroma> We decided to not have any dependency on Share or any Alfresco UI framework.

2018-02-05 11:40:11 GMT <fwu> they have some good backend. Nvertheless, Im not telling Alvex has a case system. Im ust wondering about solving the case concept as they solved the related workflow concept.

2018-02-05 11:40:30 GMT <fwu> Tichodroma: no share here also.

2018-02-05 11:40:33 GMT <fwu> custom app

2018-02-05 11:40:34 GMT <Tichodroma> :)

2018-02-05 11:41:18 GMT <Tichodroma> well, you could store information in custom tables inside Alfresco. But this would replicate work already done in CRM systems that exist and often are in use.

2018-02-05 11:41:58 GMT <Tichodroma> Connecting Alfresco to such a specialised system using an adaptor API is the more flexible and lasting approach.

2018-02-05 11:45:36 GMT <fwu> still I would need to integrate two systems. My biggest problem would be the UI

2018-02-05 11:46:26 GMT <Tichodroma> Yes. You won't use an existing CRM or Case Management system that already has an UI?

2018-02-05 11:49:46 GMT <fwu> I was trying to leverage Alfresco for that. And in that case, the UI would not be the problem. Alfresco already has the most important artfacts for a case management system: documents and workflows.

2018-02-05 11:51:10 GMT <dgradecak> fwu: I think you are not correct. The UI would be the biggest problem

2018-02-05 11:52:43 GMT <dgradecak> how I do it, when starting workflow or storing documents related to a "case", probably everyone else does the same, just store the keys that document/process are referenced in, and in the search you just use those custom key properties

2018-02-05 11:53:08 GMT <dgradecak> but still you need a UI to aggregate your documents and workflows in the smae place as your case

2018-02-05 11:53:46 GMT <hi-ko> fwu: I've seen a few aproaches to implement use cases which are already implemented in other systems in a much better way. Trying to implement these with Alfresco concepts will end up in a race between the hare and the tortoise

2018-02-05 11:55:50 GMT <hi-ko> We used the opens source CRM system EspoCRM as a reference system for cases because they already implemented the expected relations like assignee, tasks, status, team, mails, company, contact(s) ...

2018-02-05 11:57:16 GMT <hi-ko> fwu: new extensionis bring the controls and views back to Alfresco so Alfresco share looks like having all this context stored internally.

2018-02-05 11:57:17 GMT <fwu> hi-ko, I understand that. But maybe Im not looking to a so complete concept of case management.

2018-02-05 11:57:23 GMT <fwu> but what do you do regarding the UI?

2018-02-05 11:57:41 GMT <fwu> th eusers will get Alfresco UI + EspoCRM UI?

2018-02-05 11:58:52 GMT <hi-ko> basic data are shown in Alfresco using generic data controls and folder view extensions but the users have a link in every "external entity" to jump to that spedific UI

2018-02-05 12:00:05 GMT <hi-ko> So it is nearlly zero devolopment but maximum flexibility and users get the full functionallity

2018-02-05 12:12:33 GMT <fwu> hi-ko, thank you for your insights. I will have a look into EspoCRM.

2018-02-05 12:13:22 GMT <fwu> thank you also dgradecak and Tichodroma

2018-02-05 12:37:22 GMT <douglascrp> morning ugys

2018-02-05 12:37:24 GMT <douglascrp> guys

2018-02-05 12:37:44 GMT <douglascrp> I have not tried it, but how about the ootb preview in share for a 2 GB pdf file?

2018-02-05 12:37:50 GMT <douglascrp> I am going to upload it

2018-02-05 12:38:18 GMT <douglascrp> I guess it can be a problem, so I would like to antecipate on this, and ask for any advice from you if you have any to share

2018-02-05 12:40:16 GMT <Loftux1> douglascrp: There is a flag for progressive loading. We experimented with that when it was standalone project, think the flag is still there. I'll see if I can find it.

2018-02-05 12:40:53 GMT <douglascrp> Loftux1, are you talking about the pdf internal flag?

2018-02-05 12:41:08 GMT <douglascrp> or a pdfjs' flag?

2018-02-05 12:41:44 GMT <douglascrp> because I know about this, but in this case, it's dependent on the user to check that when creating the file

2018-02-05 12:42:00 GMT <douglascrp> and you know, counting on user to execute extra steps is not the best

2018-02-05 12:42:58 GMT <Loftux1> douglascrp: https://github.com/loftuxab/share-community-loftux/blob/loftux/share/src/main/webapp/components/preview/PdfJs.js#L209

2018-02-05 12:42:59 GMT <alfbot> Title: share-community-loftux/PdfJs.js at loftux · loftuxab/share-community-loftux · GitHub (at github.com)

2018-02-05 12:44:19 GMT <Loftux1> douglascrp: So it is in Alfresco standard pdfjs.js, but since Alfresco hasn't updated pdf.js core in a while I'm not sure how well it works.

2018-02-05 12:44:20 GMT <douglascrp> Loftux1, interesting

2018-02-05 12:44:29 GMT <douglascrp> tks for the tip

2018-02-05 12:44:49 GMT <Loftux1> Add it by overriding configuration for pdf viewer

2018-02-05 12:45:37 GMT <douglascrp> yes, I am going to do some quick tests

2018-02-05 12:46:58 GMT <Loftux1> I remeber when testing it that it worked ok, and Alfresco supports loading in chunks. The trick is to find chunks in the pdf that gives enough information to actually render.

2018-02-05 12:47:00 GMT <AFaust> douglascrp: But I believe you are correct - both the viewer and the PDF need to be setup for progressive viewing. If the PDF content is not streamlined, the viewer has no chance to optimise it...

2018-02-05 12:47:29 GMT <douglascrp> AFaust, yes, exactly what I thought

2018-02-05 12:47:49 GMT <douglascrp> pdfjs will try it, but it is dependent on the pdf to be able to provide it

2018-02-05 12:48:13 GMT <AFaust> Though I have not come across a PDF that was not optimised in the last couple of years - at least not that I was aware...

2018-02-05 12:49:11 GMT <douglascrp> this is for a maganize, so I guess who made it knew what was doing

2018-02-05 12:52:06 GMT <douglascrp> AFaust, well, in this case, "Optimized = false"

2018-02-05 12:53:49 GMT <AFaust> "knew what was doing" != "optimized = false"

2018-02-05 12:59:11 GMT <hi-ko> Loftux1: progressive loading requires byte range access which is not supported by Alfresco

2018-02-05 13:19:48 GMT <Loftux1> hi-ko: Maybe I remember wrong then, ti was some time ago i tested this. A quick search and there is a HttpRangeProcessor.java class.

2018-02-05 13:39:32 GMT <hi-ko> Loftux1: My tests / research may be outdated and based on 5.0 where Alfresco support told us that they don't support range requests for previews. So maybe the HttpRangeProcessor is in place but not working as expected when consumed from rendition webscripts? It's maybe worth a research.

2018-02-05 16:50:32 GMT <douglascrp> my upload has finished, but the document is not shown in the document library

2018-02-05 17:26:32 GMT *** DragiBus_ is now known as DragiBus

2018-02-05 17:38:59 GMT <hi-ko_> douglascrp: is your use case a print highres pdf?

2018-02-05 17:40:17 GMT <douglascrp> hi-ko_, no, this is just to keep the documents for a magazine

2018-02-05 17:40:23 GMT <douglascrp> the publishing documents they create

2018-02-05 17:40:27 GMT <douglascrp> usually, big files

2018-02-05 17:40:32 GMT <douglascrp> images, pdfs and videos

2018-02-05 17:42:16 GMT <hi-ko_> douglascrp: most prepress organisations have an OPI software in place which is able to switch highres and lowres images on the fly when "printing" to PDFs/Postscript device

2018-02-05 17:43:03 GMT <douglascrp> ok, so in this case, what exactly you mean:

2018-02-05 17:43:05 GMT <douglascrp> ?

2018-02-05 17:43:21 GMT <douglascrp> they should prepare the documents before sending it to alfresco?

2018-02-05 17:43:40 GMT <hi-ko_> They could create much smaller files when using the right print pipeline.

2018-02-05 17:43:43 GMT <douglascrp> or you are suggesting that I integrate such software inside alfresco to create the lowres rendition?

2018-02-05 17:43:51 GMT <douglascrp> got it

2018-02-05 17:43:59 GMT <douglascrp> this 2 GB pdf has only 40 pages

2018-02-05 17:44:05 GMT <douglascrp> I think this is too much

2018-02-05 17:45:14 GMT <hi-ko_> on the other hand: we implemented for a customer a "poor man OPI" based on alfersco using different mount points when rendering the page files.

2018-02-05 17:47:59 GMT <hi-ko_> we create a custom media management model where a media file may have different media renditions configured in global configurations. Policies take care to create jobs which queue tasks to render the media files which will be stored back in Alfresco referenced to the original file but in a separate tree.

2018-02-05 17:52:48 GMT <hi-ko_> The 2GB is not unusual for PDF/X created for printing.

2018-02-05 17:54:00 GMT <hi-ko_> but if they want to provide a pdf to the customer as a digital version the should create the pdf using other profiles instead of struggeling with that in alfresco

2018-02-05 20:57:47 GMT <yogehs> Hi everyone !! I have question on alfresco sdk 3.0.0 !! We are upgrading from 4.1.4 to 5.2.2 ! The question should we use alfresco sdk 3.0.0 for developing code which compatible with 5.2.2 !! weather alfresco sdk 3.0.0 is stable release or not ?

2018-02-05 21:01:10 GMT <yogesh> Hi everyone !! I have question on alfresco sdk 3.0.0 !! We are upgrading from 4.1.4 to 5.2.2 ! The question should we use alfresco sdk 3.0.0 for developing code which compatible with 5.2.2 !! weather alfresco sdk 3.0.0 is stable release or not ?

End of Daily Log

The other logs are at http://esplins.org/hash_alfresco