Menu
▾
▴
Re: [Exist-open] merging XML
|
From: Christian W. <cwi...@gm...> - 2021-07-27 02:12:22
|
Hi Kevin, This sounds interesting. Is you code available somewhere? I'd like to study it, especially the XQuery part. All the best, Christian On 27/07/2021 04.19, kev...@xp... wrote: > > We have a similar system for patient consent documents and several > other hospital-related documents. > > We started with a combination for the authors, some customized > Summernote (like CKEditor) and also Microsoft Word. > > The Word environment proved the best because we developed 2-way > support for exist-db. > > Outbound we build a custom “document.xml”, using XSLT converting from > their XML to Microsoft’s WORDML, stick it into a folder that has all > the other .DOCX components, zip it up and send to download as a DOCX. > All of this is in XQuery so they simply check out a document (or > create a new default one) and after a second or two, a DOCX is > downloaded to them. > > Author’s could work on what they wished and when finished upload the > DOCX as a checkin. > > We use XQuery then to unzip, extract the “document.xml” and then > basically reverse it through XSLT to make their XML and update the > database. > > This method also worked great with their language translation vendors > who run the DOCX through translation memory and upload the new > French/Arabic/Welsh/Chinese/15+ other languages. > > Word Styles are used to understand structures (headings, lists, > captions) but also impart logic. They have “chunks” of information > that are finally rendered on demand to PDF through RenderX from > exist-db (live at the hospital when needed) and the output content is > customized with Choose/When or If structures that use Word styles to > convey that. Like “If the logged in hospital is in the UK, show this > paragraph, but if in Australia show this one …” > > Anyway, we found Word to be the best authoring environment for the > authors primarily because XSLT can be used to strip out things not > needed and while it is complex, it is all there. Even with HTML > editors like CKEditor or Summernote, authors would write in Word and > then copy/paste into the editor and the results (while they look ok, > are rife with custom Word namespaces and HTML entities) that you have > to program around and/or strip Javascript side before upload. With > Word, we even extract the images (all SVG) and embed them into the > Word document so they can see them and write content referencing it > since modern Word supports SVG in Word documents now. > > Kevin > > *From:* Alfredo Cosco <alf...@gm...> > *Sent:* Monday, July 26, 2021 7:31 AM > *To:* Joe Wicentowski <jo...@gm...> > *Cc:* exist-open <exi...@li...> > *Subject:* Re: [Exist-open] merging XML > > Hi Joe, > yes I did, a couple of years ago, but I discarded the use of XForms > because for the textareas they used an editor (it seems to me > CKeditor) and to change it was necessary to completely recompile the > application (and I have no idea how to do it) . > Since I needed an XML editor I developed my own CODEMIRROR > configuration dedicated to TEI-XML which guides my editors to write > XML easily and, on the side, also has a "note taking environment" > > https://orazionelson.github.io/CodeMirrorXML4TEI/ > <https://orazionelson.github.io/CodeMirrorXML4TEI/> > https://orazionelson.github.io/CodeMirrorXML4TEI/editor.html > <https://orazionelson.github.io/CodeMirrorXML4TEI/editor.html> > > Thanks for the suggestions, I will try to study the links you gave me. > > Alfredo > > Il giorno lun 26 lug 2021 alle ore 15:28 Joe Wicentowski > <jo...@gm... <mailto:jo...@gm...>> ha scritto: > > Hi Alfredo, > > It sounds like you're trying to take the input of HTML forms and > generate XML documents from the input. Have you investigated > XForms, which can natively store XML models in the browser and > submit XML instances to eXist? For XForms implementations, see > XSLTForms [1] and Fore ("XFormish") [2]. To learn more about > XForms, see Steven Pemberton's tutorial [3]. > > [1] > https://github.com/AlainCouthures/declarative4all/raw/master/public/direct/xsltforms.zip > <https://github.com/AlainCouthures/declarative4all/raw/master/public/direct/xsltforms.zip> > > [2] https://github.com/Jinntec/Fore <https://github.com/Jinntec/Fore> > > [3] https://homepages.cwi.nl/~steven/xforms/xforms-hands-on/ > <https://homepages.cwi.nl/~steven/xforms/xforms-hands-on/> > > Joe > > On Sun, Jul 25, 2021 at 1:46 PM Alfredo Cosco > <alf...@gm... <mailto:alf...@gm...>> wrote: > > Hello Joe, Hi all, > > thanks for the code on github, I'm going to use it as a > starting point but reading your message I realized that in few > I'll have other problems to face. > > For example, if an element admits more than one attribute I'll > have two identical code snapshots to merge with different > attributes. > > I'll try to write rules from different cases, but after this > thread I'm also thinking of rewriting all the logic of my script. > > What I'm trying to do is to create a TEI document starting > from a web-form, the form may vary depending from the user > configurations > > So: > > 1) I set the attribute "name" in the html input in this way: > *fileDesc_titleStmt_title* or > *fileDesc_sourceDesc_msDesc_physDesc_objectDesc *or (for > attributes) > > *fileDesc_sourceDesc_msDesc_physDesc_objectDesc_@form* > > 2) Thanks to this forum I found a way (few weeks ago) to > transform the *name* value in an XML snapshot: > > <fileDesc> > <titleStmt> > <title/> > </titleStmt> > </fileDesc> > > 3) Then I merge the snapshots and store an empty document in > the database > > 4) Then I add the element and attribute values using the > "UPDATE INSERT" > > In theory the approach works, but in practice seems to bring > more troubles than solutions :\ > > What do you think? > > Alfredo > > Il giorno dom 25 lug 2021 alle ore 17:54 Joe Wicentowski > <jo...@gm... <mailto:jo...@gm...>> ha scritto: > > Hi Alfredo, > > Slav asks good questions about the assumptions in your > sample code. Assuming that duplicates should be merged > rather than preserved, and testing only for element names > (i.e., not examining attribute or text nodes), here's a > basic query that will merge your fragments together into a > structure approximating your target: > > https://gist.github.com/joewiz/b38279b8bd33de6949bb891370f76a04 > <https://gist.github.com/joewiz/b38279b8bd33de6949bb891370f76a04> > > Note that the order of fragments in the sequence you pass > to the merge-fragments function is important. This is > because when a new fragment is tested against the running > composite document, if an element in the fragment doesn't > match the element in the composite document, the > fragment's element is deposited first before the rest of > the composite document. This is why var3's titleStmt > element appears first in the composite document, and > var2's msIdentifier appears before var1's physDesc. > > Hopefully this gets you off to a good start and you're > able to add any necessary checks besides element name to > this to meet any other unstated assumptions / requirements. > > Joe > > On Sun, Jul 25, 2021 at 7:32 AM Слава Седов > <wet...@gm... <mailto:wet...@gm...>> wrote: > > Hi, > > If your input and output set as testcase - it is 100% > testcase or data may vary? Do you need removing any > duplicates if any or it is not possible in your input? > > With best wishes, > > Slav > > вс, 25 июл. 2021 г., 14:56 Alfredo Cosco > <alf...@gm... > <mailto:alf...@gm...>>: > > Hi Jean-Paul, > > I'm trying to merge the pieces on an "in > memory" document then to store the result in a new > document. > > But this logic can change if there is a better way. > > Thanks, > > Alfredo > > Il giorno sab 24 lug 2021 alle ore 19:16 Jean-Paul > Rehr <re...@gm... <mailto:re...@gm...>> > ha scritto: > > Alfredo, are you creating a TEI document in > memory, or are you trying to insert these into > an existing document stored in eXist-db? > > JPR > > On Sat, Jul 24, 2021 at 7:04 PM Alfredo Cosco > <alf...@gm... > <mailto:alf...@gm...>> wrote: > > Hi all, > > I'm working on eXist 4.7 > > The problem is: > > I have tre variables with XML code: > > let $var1 := <fileDesc> > > <sourceDesc> > > <msDesc> > > <physDesc> > > <objectDesc form=""/> > > </physDesc> > > </msDesc> > > </sourceDesc> > > </fileDesc> > > let var2 := <fileDesc> > > <sourceDesc> > > <msDesc> > > <msIdentifier> > > <idno/> > > </msIdentifier> > > </msDesc> > > </sourceDesc> > > </fileDesc> > > let var3 := <fileDesc> > > <titleStmt> > > <title/> > > </titleStmt> > > </fileDesc> > > How can I merge there 3 snapshots and > obtain something like: > > <fileDesc> > > <titleStmt> > > <title/> > > </titleStmt> > > <sourceDesc> > > <msDesc> > > <physDesc> > > <objectDesc form=""/> > > </physDesc> > > <msIdentifier> > > <idno/> > > </msIdentifier> > > </msDesc> > > </sourceDesc> > > </fileDesc> > > Thanks, > > Alfredo > > _______________________________________________ > Exist-open mailing list > Exi...@li... > <mailto:Exi...@li...> > https://lists.sourceforge.net/lists/listinfo/exist-open > <https://lists.sourceforge.net/lists/listinfo/exist-open> > > _______________________________________________ > Exist-open mailing list > Exi...@li... > <mailto:Exi...@li...> > https://lists.sourceforge.net/lists/listinfo/exist-open > <https://lists.sourceforge.net/lists/listinfo/exist-open> > > _______________________________________________ > Exist-open mailing list > Exi...@li... > <mailto:Exi...@li...> > https://lists.sourceforge.net/lists/listinfo/exist-open > <https://lists.sourceforge.net/lists/listinfo/exist-open> > > > > _______________________________________________ > Exist-open mailing list > Exi...@li... > https://lists.sourceforge.net/lists/listinfo/exist-open |
