Hello human, I'm @whedon, a robot that can help you with some common editorial tasks. @jameshclrk, it looks like you're currently assigned as the reviewer for this paper 🎉.

⭐️ Important ⭐️

If you haven't already, you should seriously consider unsubscribing from GitHub notifications for this (https://github.com/openjournals/joss-reviews) repository. As a reviewer, you're probably currently watching this repository which means for GitHub's default behaviour you will receive notifications (emails) for all reviews 😿

To fix this do the following two things:

1. Set yourself as 'Not watching' https://github.com/openjournals/joss-reviews:

2. You may also like to change your default settings for this watching repositories in your GitHub profile here: https://github.com/settings/notifications

For a list of things I can do to help you, just type:

@whedon commands

Attempting PDF compilation. Reticulating splines etc...

👉 Check article proof 📄 👈

I've begun my review.

In general, I think the need is clear - this package is a utility for reading PDF files in Julia, which is clearly useful to some people.

The paper is intelligible, although suffers from grammatical errors throughout that should be fixed (as does the documentation). An example of a particularly ungrammatical sentence is "Almost, all the functionality of PDF understanding is entirely written from scratch in Julia with only exception of usage of certain (de)compression codecs".

I find the third paragraph ("The APIs are developed...") confusing. What is a "native script based language"? I also don't understand the meaning of the sentence "every well-developed native library out in the market need connectors".

In "Their contribution can be viewed from the following link.", the URL of the link should be included in the paper - I don't think a JOSS paper should have embedded hyperlinks without the actual URL being visible in the document.

The paper should reference other Julia packages for handling PDF documents and explain how this package differs. In particular, I believe Taro.jl is the dominant solution at the moment.

At present, I don't think the documentation is sufficient. There are no examples or tutorials referenced in the readme or documentation. The only overview, the "API structure and design" section in https://sambitdash.github.io/PDFIO.jl/stable/#API-Structure-and-Design-1, is minimal.

The documentation, as far as I've been able to find, consists solely of a description of each of dozens of API method which has been autogenerated from the docstrings in the source code and is listed in alphabetical order. It is not reasonable to expect new users to read through that to get started, and I'm not personally willing to do so and so cannot personally test to see if this package is functional.

The package did install, although I can't personally test its functionality since the documentation is too sparse for me to understand how to use it.

Running the automated test (Pkg.test("PDFIO")) fails for me (see sambitdash/PDFIO.jl#62). From the logs, it looks like the issue is ultimately with a different downstream package (ZipFile), but nevertheless I can't verify this package passes its own tests.

In conclusion, I think major revisions to the documentation are needed, but if that can be accomplished, this software is a good fit for JOSS.

@malmaud Thanks for your review. I will update the documentation to address your concerns around the documentation. On the failing of the test cases, I guess for some reason the build step did not run when you added the package. If you can run the following command:

pkg> build PDFIO

Now, the test shall pass successfully.

OK, I just installed and tested in a fresh Julia environment and the tests pass now.

👋 @jameshclrk — We see no progress on your review checklist. Have you started on this? Do give us an updates.

@sambitdash — it looks like there remain several comments from @malmaud that need to be addressed. What's your status?

@labarba - I am going for a complete revamp of the documentation by the end of this month. Most of the suggested changes will be addressed as part of that. I will update when the task is accomplished.

@alexhanna Please follow up by email with @jameshclrk to know when we might expect their review.

@whedon remind @sambitdash in 3 weeks

Reminder set for @sambitdash in 3 weeks

@labarba Emailed @jameshclrk.

@whedon commands

Here are some things you can ask me to do:

# List Whedon's capabilities
@whedon commands

# List of editor GitHub usernames
@whedon list editors

# List of reviewers together with programming language preferences and domain expertise
@whedon list reviewers

EDITORIAL TASKS

# Compile the paper
@whedon generate pdf

# Compile the paper from alternative branch
@whedon generate pdf from branch custom-branch-name

# Ask Whedon to check the references for missing DOIs
@whedon check references

@whedon generate pdf

Attempting PDF compilation. Reticulating splines etc...

PDF failed to compile for issue #1453 with the following error:

Error reading bibliography ./paper.bib (line 14, column 1):
unexpected "u"
expecting space, ",", white space or "}"
Error running filter pandoc-citeproc:
Filter returned error status 1
Looks like we failed to compile the PDF

@whedon generate pdf

Attempting PDF compilation. Reticulating splines etc...

👉 Check article proof 📄 👈

@whedon generate pdf

Attempting PDF compilation. Reticulating splines etc...

👉 Check article proof 📄 👈

Please review with the Version: v0.1.5 of the software.

Thanks so much @malmaud. Can you confirm this by checking off the necessary boxes in your review?

done

Thanks @malmaud.

👋 @jarvist, @Datseris, @dmbates -- we're looking for a second reviewer for this piece after our first one bowed out. Could you possibly have a look at this?

Thank you for the invitation, but I am sorry, I have to decline, too many deadlines to meet.

Hi @jarvist -- are you able to review this now?

@whedon remove @jameshclrk as reviewer

OK, @jameshclrk is no longer a reviewer

Hi @jarvist -- are you able to review this now?

Certainly!

@whedon add @jarvist as reviewer

OK, @jarvist is now a reviewer

I've just been reviewing this software + paper, but I can't seem to edit the checkboxes (also, they are still allocated to James Clark. @alexhanna - is this something you can fix?

Generally I'm happy with the package... it installs and tests run. I think the latest draft of the paper looks good.

However, I still think the package README.md could be edited down to make it more clear, and make it easier for a user to understand what the functionality is, and to get started.
For instance, you can:-
julia> pdPageExtractText(stdout, pdDocGetPage(pdDocOpen("10.21105.joss.01453.pdf"), 1))
from the REPL to extract the first page of the JOSS review into the terminal. This is pretty cool functionality, but I had to find this out by fiddling around with it.

Are there any other high level access functions than pdPageExtractText ?

From the point of view of the Package/API (not having read the source code), the namespace is quite confusing: the package is PDFIO but then all the exported functions are prepended with pd which seems superfluous for Julia, because everything is already under the PDFIO namespace of the module.
Similarly, the function arguments seem to vary quite a bit in placing and expectation. Does pdPageExtractText need an io object as the first argument? A default if single argument to return the text object would make a lot of sense. (Making the stdout in my example above superfluous.)
Also, the library seems to fail on Google-docs exported PDFs. (My first test case.) I will file an issue.

Hrm, try now @jarvist? I changed the name on the issue, but I'm not sure if that does anything.

Still no 'edit' option in the '...' drop down menu.

You shouldn't need to edit it. You can click the checkboxes without editing. Does that work for you?

No joy there either. (I think the GUI checkbox clicking is predicated by edit permissions.)

Hi @jarvist,

Thanks a lot for taking time to review the software and the paper. Also reporting the bug on the product.

Are there any other high level access functions than pdPageExtractText ?

All methods starting with pd documented under: https://sambitdash.github.io/PDFIO.jl/dev/#PD-1 are high level functions.

From the point of view of the Package/API (not having read the source code), the namespace is quite confusing: the package is PDFIO but then all the exported functions are prepended with pd which seems superfluous for Julia, because everything is already under the PDFIO namespace of the module.

pd does not stand for PDFIO but PDF Document or simply called PD Layer in most PDF format architectures. This is to differentiate from low level functions called COS Layer which requires you to understand the PDF objects like dictionary, arrays to be able to understand the PDF file. An explanation of these can be seen at: https://sambitdash.github.io/PDFIO.jl/dev/arch/

Moreover, the similar nomenclature is used by other established PDF libraries in the market. Hence, people who have used other PDF libraries may quickly understand the scope of the APIs.

Similarly, the function arguments seem to vary quite a bit in placing and expectation. Does pdPageExtractText need an io object as the first argument? A default if single argument to return the text object would make a lot of sense. (Making the stdout in my example above superfluous.)

1. PDF files can be large expanding up to several megabytes and can technically grow to gigabytes. With extract text returning strings this may lead to several copies of large strings. IO provides a cleaner interface for the same. Many a times users may extract all pages and IO provides a possibility to append pages over one another.
2. PDF being an encrypted file format immutable extracted text should not be left in the memory which is a definite possibility with usage of strings. IO can be used to immediately write to a file or clear up the IOBuffer at the end of using the data.

Also, the library seems to fail on Google-docs exported PDFs. (My first test case.) I will file an issue.

This has been fixed. If you think this is a gating requirement for the paper, please let me know so that I can release a new version of the software.

👋 @openjournals/joss-eics can you look into the checkbox issue for new reviewers?

@arfon - can you help with the reviewer invitation here?

@jarvist - you've been invited by GitHub as a collaborator on this repository - you just need to accept the invite at this URL: https://github.com/openjournals/joss-reviews/invitations