aerhard / linter-autocomplete-jing Goto Github PK

[Enter steps to reproduce:]

1. ...
2. ...

Atom: 1.52.0 x64
Electron: 6.1.12
OS: Unknown Windows version
Thrown From: linter-autocomplete-jing package 0.8.4

Stack Trace

Uncaught Error: Unencoded <
Line: 1
Column: 1
Char: <

At events.js:177

Error: Unencoded <
Line: 1
Column: 1
Char: <
    at error (/packages/linter-autocomplete-jing/node_modules/sax/lib/sax.js:651:10)
    at strictFail (/packages/linter-autocomplete-jing/node_modules/sax/lib/sax.js:677:7)
    at SAXParser.write (/packages/linter-autocomplete-jing/node_modules/sax/lib/sax.js:1080:13)
    at SAXStream.write (/packages/linter-autocomplete-jing/node_modules/sax/lib/sax.js:239:18)
    at ReadableString.ondata (_stream_readable.js:713:22)
    at ReadableString.emit (events.js:200:13)
    at ReadableString.Readable.read (_stream_readable.js:508:10)
    at flow (_stream_readable.js:977:34)
    at resume_ (_stream_readable.js:958:3)
    at processTicksAndRejections (internal/process/task_queues.js:84:9)

Commands

     -7:59.7.0 core:delete (input.hidden-input)
     -7:56.8.0 core:backspace (input.hidden-input)
  3x -7:54.4.0 editor:newline (input.hidden-input)
 16x -7:47.8.0 core:backspace (input.hidden-input)
     -7:41.4.0 editor:newline (input.hidden-input)
 13x -7:40.7.0 core:backspace (input.hidden-input)
     -2:48.2.0 core:delete (input.hidden-input)
     -2:42.7.0 intentions:highlight (input.hidden-input)
     -2:42.5.0 symbols-view:toggle-file-symbols (input.hidden-input)
     -2:36.7.0 core:move-up (input.hidden-input)
     -2:36.4.0 editor:newline (input.hidden-input)
     -2:30.2.0 core:backspace (input.hidden-input)
     -2:29 autocomplete-plus:confirm (input.hidden-input)
     -2:29 snippets:expand (atom-text-editor.editor.is-focused.autocomplete-active)

Non-Core Packages

busy-signal 2.0.1 
html-tag-auto-complete 2.0.0 
intentions 1.1.5 
language-dtd 0.2.1 
linter 3.1.1 
linter-autocomplete-jing 0.8.4 
linter-ui-default 2.1.2 
linter-xmllint 1.4.4 
open-in-browser 0.5.2 
xml-formatter 0.13.0 

I need to reference a certain xsd depending on the root element used in XML.
The root element used is

To do this I've defined a rule in config.cson. However, when I load any matching XML file there's no validation or autocompletion done.

  "linter-autocomplete-jing":
    rules: [
      {
      priority: 2
      test:
        rootLocalName: "grosskommentar"
      outcome:
        schemaProps: [
          {
            lang: "xsd"
            path: "/home/holger/Dokumente/specs/SF5-Schema 0.7.1/Schema/verlag-sf.xsd"
          }
        ]
    }
]

A standard DITA XML topic usually never uses an absolute path to the grammar file, just the name of the file without a path (topic.dtd in this case). The correct grammar file is picked using the PUBLIC identifier (-//OASIS//DTD DITA Topic//EN in this case).

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE topic PUBLIC "-//OASIS//DTD DITA Topic//EN" "topic.dtd">
<topic id="topic">
    <title></title>
    <body/>
</topic>

jing complains that there is no DTD in the same directory, so it seems to not process the PUBLIC identifier. It would be nice, if it would do so.

Could not process schema or catalog: file not found: /home/stefan/Schreibtisch/topic.dtd (Datei oder Verzeichnis nicht gefunden)

Does not work

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="urn:oasis:names:tc:dita:rng:generalTask.rng" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="urn:oasis:names:tc:dita:rng:generalTask.rng" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<task id="task">
    <title>Task</title>
    <shortdesc></shortdesc>
    <taskbody>
        <context>
            <p>Context for the current task</p>
        </context>
    </taskbody>
</task>

Further on, Uniform Resource Name (URN) support would also be nice.

Works

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="/home/stefan/workspace/DITA/dita-ot/plugins/org.oasis-open.dita.v1_3/rng/technicalContent/rng/task.rng" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="/home/stefan/workspace/DITA/dita-ot/plugins/org.oasis-open.dita.v1_3/rng/technicalContent/rng/task.rng" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<task id="task">
    <title>Task</title>
    <shortdesc></shortdesc>
    <taskbody>
        <context>
            <p>Context for the current task</p>
        </context>
    </taskbody>
</task>

documents starting with <?xml version="1.0" encoding="UTF-16"?> result in the 'Content is not allowed in prolog' error message

<?xml version="1.0" encoding="utf-8"?> <geodbupdate version="1.00.00" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:noNamespaceSchemaLocation="C:\Users\NL01031\Box Sync\BC9815 Technical Data Exchange\900 Geonis Blaeu product\Versie 1.6 release\02 Software\Desktop\GEONIS Structuur\xsd\updatemodul.xsd"> ... </updateset> </geodbupdate>

results in a Jing Warning:

"Could not process schema or catalog: exception "java.io.FileNotFoundException" thrown: schema_reference.4: Failed to read schema document 'file:/C:/Users/NL01031/Box', because 1) could not find the document; 2) the document could not be read; 3) the root element of the document is not xsd:schema.: C:\Users\NL01031\Box (The system cannot find the file specified)at line 1 col 1"

Don't know if this is a linter-auto-complete issue or a jing issue.

see https://github.com/Schematron/schematron and https://github.com/dmj/schxslt

Dear @aerhard
we are using this package to validate in Atom files like this one.

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="https://raw.githubusercontent.com/BetaMasaheft/guidelines/master/schema/out/tei-betamesaheftGL.rng"
schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="https://raw.githubusercontent.com/BetaMasaheft/guidelines/master/schema/out/tei-betamesaheftGL.rng"
type="application/xml" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0">
   <teiHeader>
      <fileDesc>
         <titleStmt>
            <title>Repositories</title>
            <author>Pietro Liuzzo</author>
         </titleStmt>
         <publicationStmt>
            <p/>
          </publicationStmt>
         <sourceDesc>
            <p>Produced by pandoc. Conversion from GIT flavoured markdown to TEI simple.</p>
            <p>transformed to TEI documentation with XSLT by Pietro Liuzzo</p>
         </sourceDesc>
      </fileDesc>       <revisionDesc>       <change who="PL" when="2018-04-30">first version of guidelines from Wiki</change>
         <change who="DR" when="2018-05-15">Edited and updated example</change>
         <change who="DR" when="2019-01-10">Clarification</change> </revisionDesc>
   </teiHeader>
   <text>
      <body xml:id="institution">

         <div type="level1">
            <head>Specific encoding for repositories</head>
            <p>For repositories where manuscripts are preserved all general guidelines and especially those for <ref target="places">places</ref> are valid.
               All repository records should have a <att>subtype</att> with <val>institution</val>.</p>
            <egXML xmlns="http://www.tei-c.org/ns/Examples">
               <listPlace>
                  <place type="church dabr" subtype="institution">
                     <placeName xml:id="n1" xml:lang="gez">ዓደቃሐርሲ፡ ጰራቅሊጦስ፡</placeName>
                     <placeName corresp="#n1" xml:lang="gez" type="normalized">ʿĀddaqāḥarsi Ṗarāqliṭos</placeName>
                     <placeName type="alt" xml:id="n2">ዓደቃሐርሲ፡ መካነ፡ ህይወት፡ ጰራቅሊጦስ፡</placeName>
                     <placeName type="normalized" corresp="#n2">ʿAddaqāḥarsi Makāna Hǝywat Ṗarāqliṭos</placeName>
                     <country ref="LOC3010Ethiop"/>
                     <region notBefore="1991" type="RegionalState" ref="LOC5932Tegray"/>
                     <settlement ref="LOC3640GuloMa" type="warada" when="2010-05-03"/>
                     <settlement when="2010-05-03" ref="LOC6517Kesad" type="tabiya"/>
                     <settlement when="2010-05-03" ref="LOC6518BetaP" type="qushat"/>
                     <location>
                        <geo>14.308833 39.378683</geo>
                        <height unit="m">2437.0</height>
                     </location>
                     <note>
                        <date type="foundation" notBefore="1400" notAfter="1500">1400-1500</date>
                        <desc type="foundation">Existed aready in the pre-Christian time as the religious site of South Arabian migrants; the church was founded in a later period.</desc>
                       
                     </note>
                     <listBibl type="secondary">
                        <bibl>
                           <ptr target="bm:Nosnitsin2014Addaqaharsi"/>
                        </bibl>
                        <bibl>
                           <ptr target="bm:Nosnitsin2013Churches"/>
                        </bibl>
                     </listBibl>
                  </place>
               </listPlace>
</egXML>

            <p>See also <ref target="placeOrInstitution">Place or Institution</ref>.</p>
         </div>


      </body>
   </text>
</TEI>

It has always worked like a charm until the latest atom and package update. All seams to install well, but I get no warnings or autocompletion at all any more.
Can you help me to understand were the problem is? do we need to set things differently now?

Hello,

Hello @aerhard ,

When using the autocomplete functionality, for each element I create, the attribute xmlns="null" is added (excerpts of xml and xsd below). I don't know enough about these kind of things to suspect if this is an issue with the xml, underlying xsd or the auto-completer. I know enough that it is annoying and it would help if this would not happen (perhaps as an option).

(If something is clearly wrong with either the xml or xsd, I'm really sorry for bothering you with my stupidity :)

xml:
<?xml version="1.0" encoding="UTF-8"?>
<mnu version="5.2.0" xsi:noNamespaceSchemaLocation="..\..\..\xsd_520\mnu.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
...
</mnu>

xsd:
<?xml version="1.0" encoding="UTF-8"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema">
...
</xs:schema>

I am new to this. I have followed the installation instructions, but currently I repeatedly get the error in the image attached. I have checked the path for Java and it is correct. Can you advise?

< renders as <, & as & etc.

This is not a bug, but a confusion. I'm getting different behaviors dealing with xinclude when validating the same xml file with the same schema in Oxygen and in Atom.

Might you try this file on your own system?

https://bitbucket.org/jeffreycwitt/ppr-lectio17/raw/996460a757cdaf88dabef234af5247ebc6bacb1a/pal_ppr-lectio17.xml

In Oxygen there is no error.

But in Atom I get this error:

RNG element "xi:include" not allowed here; expected the element end-tag or element "ab", "addSpan", "alt", "altGrp", "anchor", "app", "argument", "bibl", "biblFull", "biblStruct", "byline", "cb", "cit", "damageSpan", "dateline", "delSpan", "desc", "div", "divGen", "docAuthor", "docDate", "epigraph", "floatingText", "fw", "gap", "gb", "head", "index", "interp", "interpGrp", "join", "joinGrp", "l", "label", "lb", "lg", "link", "linkGrp", "list", "listApp", "listBibl", "listEvent", "listNym", "listOrg", "listPerson", "listPlace", "listTranspose", "listWit", "meeting", "metamark", "milestone", "msDesc", "note", "opener", "p", "pb", "q", "quote", "said", "salute", "signed", "sp", "space", "span", "spanGrp", "stage", "substJoin", "timeline" or "witDetail"

Remove the schema declaration and the error goes away.

However, again, in Oxygen the document validates with the schema declaration and xinclude present? Any ideas what's going on?

BTW, It's a fantastic plug-in. And if I can fix this error, I plan to use it heavily! Where's your "donate" button?

Hi Alexander!

First of all, thank you for creating such a useful plugin!
I'm always on the lookout for tools that allow me to properly validate xml messages, without requiring me to fire up XMLSpy.

That said, I've bee able to configure my atom config.cson in order to add a rule for a specific schema like this:

    rules: [
      {
        priority: 1
        test:
          rootNs: "urn:namespace1:7:0"
        outcome:
          schemaProps: [
            {
              lang: "xsd"
              path: "D:\\borrar\\xsd\\namespace1_v7_0.xsd"
            }
          ]
      }
    ]

This works like a charm!

However, I'd like to be able to validate against many different schemas. Instead of entering each schema in the config, I wanted to have a more portable solution to my namespace-schema mappings. I thought using a catalog would do the trick. But when I try to use a catalog in the rules like this:

      {
        priority: 1
        test:
          rootNs: "urn:namespace2:7:1"
        outcome:
          xmlCatalog: "D:\\borrar\\xsd\\xbid-catalog.xml"
      }

The content of the catalog:

<?xml version="1.0"?>
  <catalog xmlns="urn:oasis:names:tc:entity:xmlns:xml:catalog">

    <system systemId="urn:namespace1:7:0"
            uri="namespace1_v7_0.xsd"/>

    <system systemId="urn:namespace2:7:1"
            uri="namespace2_v7_1.xsd"/>

  </catalog>

When I load an xml file with a namespace in the catalog I get an error

[linter-autocomplete-jing] Cannot convert undefined or null to object
TypeError: Cannot convert undefined or null to object
    at Function.from (native)
    at toConsumableArray (file:///d:/Usuarios/marbarag/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1:1)
    at file:///d:/Usuarios/marbarag/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1:1
    at getSchemaProps (file:///d:/Usuarios/marbarag/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1:1)
    at Object.lint (file:///d:/Usuarios/marbarag/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1:1)
    at promises.push.Promise.then._this.emitter.emit.linter (file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/linter-registry.js:63:26)
    at file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/linter-registry.js:62:23
    at Set.forEach (native)
    at LinterRegistry.lint (file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/linter-registry.js:60:18)
    at file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/linter.coffee:98:16
    at Function.module.exports.Emitter.simpleDispatch (D:\Usuarios\marbarag\AppData\Local\atom\app-1.12.6\resources\app.asar\node_modules\event-kit\lib\emitter.js:25:14)
    at Emitter.module.exports.Emitter.emit (D:\Usuarios\marbarag\AppData\Local\atom\app-1.12.6\resources\app.asar\node_modules\event-kit\lib\emitter.js:129:28)
    at file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/editor-linter.js:86:22
    at file:///d:/Usuarios/marbarag/.atom/packages/linter/lib/helpers.coffee:25:18

Maybe I'm trying to use catalogs for something they're not for (I don't want to add any explicit schemaLocation references to my xml files).

Do you have any recommendation on how to configure many namespace-schema mappings outside of atom's config and still use your plugin?

Thank you!!!

Thanks for the great plugin! 🎉

At https://openstax.org we edit XML files that match a RNG schema but unfortunately we do not provide a schema URL in those files. Instead, we rely on a file extension (*.cnxml).

Could there be a setting that allows the plugin to resolve a default schema if none is provided in the file? Maybe a setting that allows a text field like *.cnxml|rng|/path/to/file.rng *.foo.xml|rng|/path/to/another.rng?

I started looking around the code and it seems that getSchemaProps.js could parse that setting and append it to schemaProps if textEditor.getFileName() matches one of the glob patterns.

Something like that would really help us out; do you think it would be a useful feature to add?

ID constraint checks cause a NullPointerException, parser exits too early and suggestions get output for the element causing the exception instead of the descendant / following element at the cursor

Don't think this is local to my setup - updated atom, and got this deprecation warning:

Argument to path.dirname must be a string

Object.dirname (/usr/share/atom/resources/app/src/electron-shims.js:9:10)
parseRule (/home/reedstrm/.atom/packages/linter-autocomplete-jing/lib/main.coffee:3230:23)
arrayMap (/home/reedstrm/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1552:21)
map$1 (/home/reedstrm/.atom/packages/linter-autocomplete-jing/lib/main.coffee:1697:10)
<unknown> (/home/reedstrm/.atom/packages/linter-autocomplete-jing/lib/main.coffee:2314:12)
<unknown> (/home/reedstrm/.atom/packages/linter-autocomplete-jing/lib/main.coffee:2214:42)

We have a complex relaxng schema, that has patterns like this, where we're conditioning content within an attribute value based on other attributes.

To do that, we have multiple places where we define the same attribute name, on the same element.

What I expect: auto-completion only lists an attribute suggestion once for any given element.

What I see: duplicate attribute suggestions (see below).

Suggested fix: add code to ensure an attribute suggestion only outputs once per element.

Wrong line is 4. But message is Line 1:1. How fix it?

Could not process schema or catalog: exception "com.thaiopensource.resolver.ResolverException" thrown: java.net.URISyntaxException: Illegal character in opaque part at index 2: C:\Users\Eduardo Gutierrez\Desktop\Correcciones\M4 UF1 RA2RA3 �?? Practica EAF4\Practica4\Exercici4.xml

XML file:
`

   <película código="p17" duración="06">
         <título>EL ORO DE MOSCÚ</título>
         <nacionalidad>España</nacionalidad>
            <nacionalidad>España</nacionalidad>
               <nacionalidad>España</nacionalidad>
         <género>Comedia</género>
         <sin_clasificar/>
         <sinopsis>
              Por una extraña coincidencia del destino, alguien recibe una información extraconfidencial de un anciano en sus últimos segundos de vida: el secreto mejor
              guardado de la Historia. El receptor, un trabajador de hospital, se lo comunica
              secretamente a un supuesto amigo. Ambos inician una aventura rocambolesca y llena
              de misterio. Ante la inutilidad de sus intentos y muy a su pesar, tienen que recurrir a otras personas que así mismo van cayendo en el pozo sin fondo que conlleva
              descifrar el enigma.
         </sinopsis>
         <director>Jesús Bonilla</director>
         <reparto>
               <actor>Jesús Bonilla</actor>
               <actor>Santiago Segura</actor>
               <actor>Alfredo Landa</actor>
               <actor>Concha Velasco</actor>
               <actor>Antonio Resines</actor>
               <actor>Gabino Diego, María Barranco</actor>
               <actor>María Barranco</actor>
         </reparto>
   </película>

Thanks

We seem to once again experience the issue described here

#58

(same project, same schema issues)

On the computer where I have version 0.8.4 all seems to work as fine as always

but on the computers where the latest version has been installed

the autocomplete and validation do not work.

Is it possible to get a work around? Or should we downgrade to 0.8.4?

Thank you.

I have the following entry in the config.cson:

{
  priority: 2
  test:
    publicId: "-//OASIS//DTD DITA Topic//EN"
    outcome:
      schemaProps: [
        {
           path: "/home/stefan/workspace/DITA/dita-ot/plugins/org.oasis-open.dita.v1_3/dtd/technicalContent/dtd/topic.dtd"
           xmlCatalog: "/home/stefan/workspace/DITA/dita-ot/plugins/org.oasis-open.dita.v1_3/dtd/catalog.xml"
        }
      ]
}

I try to validate a standard DITA-XML Topic (DTD):

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE topic PUBLIC "-//OASIS//DTD DITA Topic//EN" "topic.dtd">
<topic id="topic">
    <title></title>
    <body>
        <p></p>
    </body>
</topic>

I get the validation message:

Validation failed: no DTD found ! (DTD)	3:18
Warning	Jing	Could not process schema or catalog: file not found: /home/stefan/Schreibtisch/topic.dtd (Datei oder Verzeichnis nicht gefunden)

The validator tries to load the DTD from the same directory, where the DITA-XML file is saved, but it should use the DTD defined in config.cson. How can I fix this?

I have a package built on top of your linter-autocomplete-jing package and my package has stopped working on Atom 1.32.2 (speedata/atom-schema-speedata#5). Now I assume an error on my side or some changes in Atom, so I'd like to ask if you have heard of similar issues which could help me tracing the problem. BTW: is there an easy way to test your package?

A colleague using this package to validate XML for our project have for some time been experiencing stray characters appearing in edits using Atom.

They are seeing the following error

which looks like the validation is happening on partial element names only or something similar

Do you have any idea what may be the issue?

thank you very much

I cannot get the catalog processing working. The following file (carMaintenance.dita) throws a Could not process schema or catalog error. The DITA grammars can be found here for testing.

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="urn:oasis:names:tc:dita:rng:task.rng" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="urn:oasis:names:tc:dita:rng:task.rng" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<task id="carMaintenance">
    <title>Car Maintenance</title>
    <shortdesc>Avoid costly repairs by constantly performing simple maintenance tasks.</shortdesc>
    <taskbody/>
</task>

This is my config.cson:

"*":
  core:
    telemetryConsent: "limited"
  "exception-reporting":
    userId: "3701b290-4444-420b-9784-a1192315e69b"
  "linter-autocomplete-jing":
    displaySchemaWarnings: true
    rules: [
      {
        priority: 1
        test:
          rootNs: "urn:oasis:names:tc:dita:rng:task.rng"
        outcome:
          xmlCatalog: "/home/stefan/workspace/DITA/dita-ot/plugins/org.oasis-open.dita.v1_3/rng/technicalContent/catalog.xml"
      }
    ]
  "linter-ui-default": {}

when XML models with a Schematron schematypens reference an RNG or XSD file with embedded Schematron rules, the linter should validate against the embedded Schematron rules

Please implement support for XML catalogs.

attribute name suggestions should not only appear at <el [cursor]> but also at <el [cursor]