Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Serialization changed #34

Closed
koppor opened this issue Nov 20, 2015 · 10 comments
Closed

Serialization changed #34

koppor opened this issue Nov 20, 2015 · 10 comments

Comments

@koppor
Copy link
Member

koppor commented Nov 20, 2015

The recent master branch changes the order of fields. Is it possible to keep the order of the fields as they were in the version of october? I have not changed my preferences, it is still Save fields sorted in alphabetic order (as in versions 2.10+). @stefan-kolb maybe?

grabbed_20151120-183827

grabbed_20151120-183849

grabbed_20151120-183800

@koppor
Copy link
Member Author

koppor commented Nov 23, 2015

Well, maybe some fields moved from "required" to "optional". But Why is "Journal" getting optional in @Article? - Maybe the old ordering was wrong. I Wonder why "Year" was written before "Editor" in my @Inproceedings.

grabbed_20151123-015254

grabbed_20151123-015305

@lenhard
Copy link
Member

lenhard commented Nov 23, 2015

No, it is not possible to leave the order of fields as it was in October, since we decided during the last dev call to change it to the ordering stated in the Bibtex manual.

But as far as I know, no one deliberately changed field order and the one above also does not make sense, so this is a bug. Serialization depends on the order in which fields are stated, so the changes made here JabRef#352 are probably the cause. But at the moment I don't understand how.

@koppor
Copy link
Member Author

koppor commented Nov 23, 2015

Ordering: Yes. But I would have expected that to happen only if I write new entries and the old ones should remain untouched. This eases parallel usage of JabRef 2.10, 2.11., and 3.0.

We sort the req and opt fields when writing. Maybe there something goes wrong, because we return umodifiable lists now at the types.

@lenhard
Copy link
Member

lenhard commented Nov 23, 2015

That is true. But not modifiying existing entries is a mountain of work to implement, so I guess that no one will touch that for a very long time, if ever.

Regarding the unmodifiable lists: That change is ages old. Thus it is highly unlikely that it is the cause. I would rather guess that a resorting of the fields in the collections (when creating the types) is the reason.

@stefan-kolb
Copy link
Member

Looks like this is caused by the method getRequiredFields(). getRequiredFields() returns required fields including alternatives, e.g. author/editor. This is why it might be serialzed last. The new method getRequiredFieldsFlat() now returns all fields without alternative syntax. I thought this was replaced correctly everywhere, but we might have to keep an eye on this!

@stefan-kolb
Copy link
Member

JabRef@db15dc2

@koppor
Copy link
Member Author

koppor commented Nov 23, 2015

Works for me 👍

@koppor koppor closed this as completed Nov 23, 2015
@koppor
Copy link
Member Author

koppor commented Nov 24, 2015

@stefan-kolb Could you fix that for biblatex mode, too?

@koppor koppor reopened this Nov 24, 2015
@stefan-kolb
Copy link
Member

Is this an issue? Biblatex entries should be written like Bibtex entries, therefore this should already work as expected.

@koppor
Copy link
Member Author

koppor commented Nov 25, 2015

Think, I used a wrong jar, even if I double checked. On my second machine (Win 8.1), it works perfectly (besides the changed serialization). Java 1.8.0_66. Will recheck the next days on my primary machine.

@koppor koppor closed this as completed Nov 25, 2015
koppor pushed a commit that referenced this issue Nov 10, 2021
…idth

* upstream/main:
  Remove linting exceptions not required any more (and fix blank lines in REAMDE.md)
  Disable markdownlint for docs, because Gitbook does not produce "correct" Markdown
  Disable IacrEprintFetcherTest, because gets blocked
  Fix checkstyle
  Add empty line to have log file rendered properly
  GitBook: [#34] Add link to config of "Auto import feature"
  Adapt test to new behavior
  Add PR reference
  Remove semantic duplicate class Eprint
  Fix ArXiVIdentifier -- dot is required and not arbitrary character
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants