changes for ZNTA-1847,48,49,50

[yu-shao-gm]

https://zanata.atlassian.net/browse/ZNTA-1847
https://zanata.atlassian.net/browse/ZNTA-1848
https://zanata.atlassian.net/browse/ZNTA-1849
https://zanata.atlassian.net/browse/ZNTA-1850

[yu-shao-gm]

I made a pull request against devel branch, but looked at changes, it seems wrong?

[huangp]

@yu-shao-gm I changed it to target master.

[huangp]

I am fine with the changes. @seanf do you want to have another look?

[seanf]

Review status: 0 of 3 files reviewed at latest revision, 3 unresolved discussions, some commit checks broke.

---

src/main/java/org/jenkinsci/plugins/zanata/ZanataCliBuilder.java, line 79 at r1 (raw file):

    private final String commandG2Z;
    private final boolean syncZ2git;
    private final String commandZ2G;

@yu-shao-gm
These field names syncG2zanata, commandG2Z, syncZ2git, commandZ2G seem a bit too cryptic for long-term storage. Am I correct in assuming that these values will be stored in Jenkins' XML configuration files with these field/property names as the keys?

---

src/main/java/org/jenkinsci/plugins/zanata/ZanataCliBuilder.java, line 121 at r1 (raw file):

    }

    public String getCommandZ2G() {

I think all these properties should have javadoc comments explaining what they mean.

---

src/main/resources/org/jenkinsci/plugins/zanata/ZanataCliBuilder/config.jelly, line 28 at r1 (raw file):

  </f:entry>
  <f:entry title="Command to sync from Zanata to source Git repo" field="commandZ2G">
    <f:textarea default="find . -type f -not -path &quot;*/target/*&quot; -name 'zanata.xml' \&#10;  -execdir echo &quot;=== Commiting new translation $BUILD_TAG...\n&quot; \; \&#10;  -execdir pwd \; \&#10;  -execdir ls \; \&#10;  -execdir echo '=== Git add...\n' \; \&#10;  -execdir git add . \; \&#10;  -execdir echo '=== Git commit ...\n' \; \&#10;  -execdir git commit -m &quot;$BUILD_TAG&quot; \; \&#10;  -execdir echo &quot;=== Finished commit preparation - $BUILD_TAG.\n&quot; \;"></f:textarea>

@yu-shao-gm
These shell scripts are virtually unreadable (and unreviewable) jammed into XML attributes, and will be difficult to maintain as well.

I think it's better to get them from Java strings (as they were in DescriptorImpl), which were at least broken up into multiple lines. It would be even better to read them from resource files on the classpath.

---

Comments from Reviewable

[yu-shao-gm]

Patrick has implemented the jelly default value through Java strings. I have ported the code and I will commit next.
For variable names like commandG2Z, do we want to change to commandGitToZanata?
Anyway, I think I could change the variable names and also add Javadocs.

[yu-shao-gm]

Hi Patrick and Sean, I committed some changes to yshao-cli-changes branch, please have a look. Hope the issues here are addressed fully, and I tested ok.

[seanf]

---

Reviewed 1 of 3 files at r1, 2 of 2 files at r2.
Review status: all files reviewed at latest revision, all discussions resolved, some commit checks broke.

---

Comments from Reviewable

[seanf]

@yu-shao-gm This is the point in the process (for zanata-platform) where we generally use the "on QA" label and get a QE to try it out.

@djansen-redhat Have you got time to look at this?