Merge pull request #15 from mseri/process-file

Add file processing support

.github/workflows/workflow.yml

@@ -55,7 +55,7 @@ jobs:
       - name: Use OCaml ${{ matrix.ocaml-version }}
         uses: ocaml/setup-ocaml@v2
         with:
-          ocaml-version: ${{ matrix.ocaml-version }}
+          ocaml-compiler: ${{ matrix.ocaml-version }}
           dune-cache: true
 
       - name: Set git user
@@ -90,6 +90,7 @@ jobs:
           opam exec -- dune build @fmt
       
       - name: Run tests
+        if: ${{ !startsWith(github.ref, 'refs/tags/') }}
         run: opam exec -- dune runtest
 
       - name: Build

.ocamlformat

@@ -0,0 +1 @@
+ocaml-version = 4.08

CHANGES.md

@@ -1,3 +1,8 @@
+# 0.6.0 (2022-02-03)
+
+- Support batch processing of files of IDs
+- Support append result to file
+
 # 0.5.2 (2021-12-17)
 
 - Move from cuz to the published clz

README.md

@@ -9,44 +9,53 @@ Usage:
        arXiv or PubMed ID.
 
     SYNOPSIS
-        doi2bib [OPTION]... [ID]
+       doi2bib [OPTION]... [ID]
 
     ARGUMENTS
-        ID  A DOI, an arXiv ID or a PubMed ID. The tool tries to automatically
-            infer what kind of ID you are using. You can force the cli to
-            lookup a DOI by using the form 'doi:ID' or an arXiv ID by using
-            the form 'arXiv:ID'. PubMed IDs always start with 'PMC'.
+       ID  A DOI, an arXiv ID or a PubMed ID. The tool tries to automatically
+           infer what kind of ID you are using. You can force the cli to
+           lookup a DOI by using the form 'doi:ID' or an arXiv ID by using
+           the form 'arXiv:ID'. PubMed IDs always start with 'PMC'.
 
     OPTIONS
-        --help[=FMT] (default=auto)
-            Show this help in format FMT. The value FMT must be one of `auto',
-            `pager', `groff' or `plain'. With `auto', the format is `pager` or
-            `plain' whenever the TERM env var is `dumb' or undefined.
+       -f FILE, --file=FILE
+           With this flag, the tool reads the file and process its lines
+           sequentially, treating them as DOIs, arXiv IDs or PubMedIDs.
+           Errors will be printed on standard error but will not terminate
+           the operation.
 
-        --version
-            Show version information.
+       --help[=FMT] (default=auto)
+           Show this help in format FMT. The value FMT must be one of `auto',
+           `pager', `groff' or `plain'. With `auto', the format is `pager` or
+           `plain' whenever the TERM env var is `dumb' or undefined.
+
+       -o OUTPUT, --output=OUTPUT (absent=stdout)
+           Append the bibtex output to the specified file. It will create the
+           file if it does not exist.
+
+       --version
+           Show version information.
 
     EXIT STATUS
-        doi2bib exits with the following status:
+       doi2bib exits with the following status:
 
-        0   on success.
+       0   on success.
 
-        124 on command line parsing errors.
+       124 on command line parsing errors.
 
-        125 on unexpected internal errors (bugs).
+       125 on unexpected internal errors (bugs).
 
     BUGS
-        Report bugs to https://github.com/mseri/doi2bib/issues
-
+       Report bugs to https://github.com/mseri/doi2bib/issues
 
-It will output the bibtex entry, using the published details when possible.
+It will retrieve the bibtex entry, using the published details when possible.
 
 Examples of use (the bibtex entry is printed on standard output):
 
     $ doi2bib 10.1007/s10569-019-9946-9
-    $ doi2bib doi:10.4171/JST/226
-    $ doi2bib arXiv:1609.01724
+    $ doi2bib doi:10.4171/JST/226 -o "bibliography.bib"
     $ doi2bib 1902.00436
+    $ doi2bib arXiv:1609.01724
     $ doi2bib PMC2883744
 
 Each release comes with attached binaries for windows, mac and linux.

bin/doi2bib.ml

@@ -2,12 +2,52 @@ open Doi2bib
 
 let err s = `Error (false, s)
 
-let doi2bib id =
-  match id with
-  | None -> `Help (`Pager, None)
-  | Some id -> (
-      match Lwt_main.run (Http.get_bib_entry @@ Parser.parse_id id) with
-      | bibtex -> `Ok (Printf.printf "%s" bibtex)
+let process_id outfile id =
+  let open Lwt.Syntax in
+  let* bibtex = Http.get_bib_entry @@ Parser.parse_id id in
+  match outfile with
+  | "stdout" -> Lwt_io.print bibtex
+  | outfile ->
+      let flags = [ Unix.O_WRONLY; O_APPEND; O_CREAT ] in
+      Lwt_io.with_file ~mode:Output ~flags outfile (fun oc ->
+          Lwt_io.write_line oc bibtex)
+
+let process_file outfile infile =
+  let open Lwt.Syntax in
+  let write_out f =
+    match outfile with
+    | "stdout" -> f Lwt_io.stdout
+    | outfile ->
+        let flags = [ Unix.O_WRONLY; O_APPEND; O_CREAT ] in
+        Lwt_io.with_file ~mode:Output ~flags outfile f
+  in
+  let lines ic =
+    Lwt_seq.unfold_lwt
+      (fun ic ->
+        let* line = Lwt_io.read_line_opt ic in
+        Lwt.return @@ Option.map (fun x -> (x, ic)) line)
+      ic
+  in
+  let process oc id =
+    match Http.get_bib_entry @@ Parser.parse_id id with
+    | bibtex ->
+        let* bibtex = bibtex in
+        Lwt_io.write_line oc bibtex
+    | exception e -> Lwt_io.eprintf "Error for %s: %s" id (Printexc.to_string e)
+  in
+  Lwt_io.with_file ~mode:Input infile (fun ic ->
+      write_out (fun oc -> Lwt_seq.iter_s (process oc) (lines ic)))
+
+let doi2bib id file outfile =
+  match (id, file) with
+  | None, "" -> `Help (`Pager, None)
+  | None, infile -> (
+      match Lwt_main.run (process_file outfile infile) with
+      | () -> `Ok ()
+      | exception e -> err @@ Printexc.to_string e)
+  | Some id, "" -> (
+      match Lwt_main.run (process_id outfile id) with
+      | () -> `Ok ()
       | exception Http.PubMed_DOI_not_found ->
           err @@ Printf.sprintf "Error: unable to find a DOI entry for %s.\n" id
       | exception Http.Entry_not_found ->
@@ -29,9 +69,26 @@ let doi2bib id =
                 You can force me to consider it by prepending 'doi:', 'arxiv:' \
                 or 'PMC' as appropriate."
                id)
+  | Some _, _ -> `Help (`Pager, None)
 
 let () =
   let open Cmdliner in
+  let file =
+    let doc =
+      "With this flag, the tool reads the file and process its lines \
+       sequentially, treating them as DOIs, arXiv IDs or PubMedIDs. Errors \
+       will be printed on standard error but will not terminate the operation."
+    in
+    Arg.(value & opt string "" & info [ "f"; "file" ] ~docv:"FILE" ~doc)
+  in
+  let out =
+    let doc =
+      "Append the bibtex output to the specified file. It will create the file \
+       if it does not exist."
+    in
+    Arg.(
+      value & opt string "stdout" & info [ "o"; "output" ] ~docv:"OUTPUT" ~doc)
+  in
   let id =
     let doc =
       "A DOI, an arXiv ID or a PubMed ID. The tool tries to automatically \
@@ -42,7 +99,7 @@ let () =
     in
     Arg.(value & pos 0 (some string) None & info ~docv:"ID" ~doc [])
   in
-  let doi2bib_t = Term.(ret (const doi2bib $ id)) in
+  let doi2bib_t = Term.(ret (const doi2bib $ id $ file $ out)) in
   let info =
     let doc =
       "A little CLI tool to get the bibtex entry for a given DOI, arXiv or \

bin/dune

@@ -1,5 +1,5 @@
 (executable
  (name doi2bib)
  (public_name doi2bib)
- (libraries cmdliner doi2bib)
+ (libraries cmdliner doi2bib unix)
  (preprocess future_syntax))