Incorporate job schema generation into generate_auto_job

5 files changed, 459 insertions, 7 deletions

diff --git a/README-maintainer b/README-maintainer
index d5801070..7ea049dc 100644
--- a/README-maintainer
+++ b/README-maintainer
@@ -128,7 +128,10 @@ Command-line arguments are closely coupled with QPDFJob. To add a new
 command-line argument, add the option to the appropriate table in
 job.yml. This will automatically declare a method in the private
 ArgParser class in QPDFJob_argv.cc which you have to implement. The
-implementation should make calls to methods in QPDFJob.
+implementation should make calls to methods in QPDFJob. Then, add the
+same option to either the no-json section of job.yml if it is to be
+excluded from the job json structure, or add it under the json
+structure to the place where it should appear in the json structure.
 
 The build will fail until the new option is documented in
 manual/cli.rst. To do that, create documentation for the option by
diff --git a/generate_auto_job b/generate_auto_job
index 79abc8b9..1706ff04 100755
--- a/generate_auto_job
+++ b/generate_auto_job
@@ -5,6 +5,7 @@ import argparse
 import hashlib
 import re
 import yaml
+import json
 
 whoami = os.path.basename(sys.argv[0])
 BANNER = f'''//
@@ -29,6 +30,7 @@ class Main:
         'decl': 'libqpdf/qpdf/auto_job_decl.hh',
         'init': 'libqpdf/qpdf/auto_job_init.hh',
         'help': 'libqpdf/qpdf/auto_job_help.hh',
+        'schema': 'libqpdf/qpdf/auto_job_schema.hh',
     }
     SUMS = 'job.sums'
 
@@ -199,6 +201,9 @@ class Main:
                         raise Exception(
                             f'help for unknown option {option},'
                             f' lineno={lineno}')
+                    if option not in self.help_options:
+                        # QXXXQ also need to exclude help table
+                        self.jdata[option[2:]]['help'] = short_text
                     print(f'ap.addOptionHelp("{option}", "{topic}",'
                           f' "{short_text}", R"({long_text})");', file=f)
                     help_lines += 1
@@ -226,9 +231,12 @@ class Main:
         with open('job.yml', 'r') as f:
             data = yaml.safe_load(f.read())
         self.validate(data)
-        self.options_without_help = set(
+        # Add the built-in help options to tables that we populate as
+        # we read job.yml since we won't encounter these in job.yml
+        self.help_options = set(
             ['--completion-bash', '--completion-zsh', '--help']
         )
+        self.options_without_help = set(self.help_options)
         self.prepare(data)
         with open(self.DESTS['decl'], 'w') as f:
             print(BANNER, file=f)
@@ -242,6 +250,11 @@ class Main:
             with open('manual/cli.rst', 'r') as df:
                 print(BANNER, file=f)
                 self.generate_doc(df, f)
+        self.generate_schema(data)
+        with open(self.DESTS['schema'], 'w') as f:
+            print('static constexpr char const* JOB_SCHEMA_DATA = R"(' +
+                  json.dumps(self.schema, indent=2, separators=(',', ': ')) +
+                  ')";', file=f)
 
         # Update hashes last to ensure that this will be rerun in the
         # event of a failure.
@@ -251,6 +264,24 @@ class Main:
     def prepare(self, data):
         self.decls = []
         self.init = []
+        self.jdata = {
+            # option: {
+            #   tables: set(),
+            #   help: string,
+            #   QXXXQ something for registering handler
+            # }
+        }
+
+        def add_jdata(flag, table):
+            nonlocal self
+            if table == 'help':
+                self.help_options.add(f'--{flag}')
+            elif flag in self.jdata:
+                self.jdata[flag]['tables'].add(table)
+            else:
+                self.jdata[flag] = {
+                    'tables': set([table]),
+                }
 
         self.init.append('auto b = [this](void (ArgParser::*f)()) {')
         self.init.append('    return QPDFArgParser::bindBare(f, this);')
@@ -275,7 +306,7 @@ class Main:
         self.decls.append('')
         for o in data['options']:
             table = o['table']
-
+            table_prefix = o.get('prefix', table)
             if table == 'main':
                 self.init.append('this->ap.selectMainOptionTable();')
             elif table == 'help':
@@ -296,12 +327,14 @@ class Main:
                 self.decls.append(f'void {identifier}();')
                 self.init.append(f'this->ap.addBare("{i}", '
                                  f'b(&ArgParser::{identifier}));')
+                add_jdata(i, table_prefix)
             for i in o.get('optional_parameter', []):
                 self.options_without_help.add(f'--{i}')
                 identifier = self.to_identifier(i, prefix, False)
                 self.decls.append(f'void {identifier}(char *);')
                 self.init.append(f'this->ap.addOptionalParameter("{i}", '
                                  f'p(&ArgParser::{identifier}));')
+                add_jdata(i, table_prefix)
             for i, v in o.get('required_parameter', {}).items():
                 self.options_without_help.add(f'--{i}')
                 identifier = self.to_identifier(i, prefix, False)
@@ -309,6 +342,7 @@ class Main:
                 self.init.append(f'this->ap.addRequiredParameter("{i}", '
                                  f'p(&ArgParser::{identifier})'
                                  f', "{v}");')
+                add_jdata(i, table_prefix)
             for i, v in o.get('required_choices', {}).items():
                 self.options_without_help.add(f'--{i}')
                 identifier = self.to_identifier(i, prefix, False)
@@ -316,6 +350,7 @@ class Main:
                 self.init.append(f'this->ap.addChoices("{i}", '
                                  f'p(&ArgParser::{identifier})'
                                  f', true, {v}_choices);')
+                add_jdata(i, table_prefix)
             for i, v in o.get('optional_choices', {}).items():
                 self.options_without_help.add(f'--{i}')
                 identifier = self.to_identifier(i, prefix, False)
@@ -323,11 +358,13 @@ class Main:
                 self.init.append(f'this->ap.addChoices("{i}", '
                                  f'p(&ArgParser::{identifier})'
                                  f', false, {v}_choices);')
+                add_jdata(i, table_prefix)
             if table not in ('main', 'help'):
                 identifier = self.to_identifier(table, 'argEnd', False)
                 self.decls.append(f'void {identifier}();')
         for o in data['options']:
             table = o['table']
+            table_prefix = o.get('prefix', table)
             if 'from_table' not in o:
                 continue
             if table == 'main':
@@ -341,6 +378,79 @@ class Main:
             for j in ft['options']:
                 self.init.append('this->ap.copyFromOtherTable'
                                  f'("{j}", "{other_table}");')
+                add_jdata(j, table_prefix)
+
+    def generate_schema(self, data):
+        # XXX check data['json'] against what we know from jdata.
+        # Ultimately be able to generate a schema as well as
+        # JSONHandler and registering stuff.
+
+        # Check to make sure that every command-line option is
+        # represented either in data['json'] or data['no-json'].
+
+        # Build a list of options that we expect. If an option appears
+        # once, we just expect to see it once. If it appears in more
+        # than one options table, we need to see a separate version of
+        # it for each option table. It is represented prepended in
+        # job.yml with the table prefix. The table prefix is removed
+        # in the schema.
+        expected = {}
+        for k, v in self.jdata.items():
+            tables = v['tables']
+            if len(tables) == 1:
+                expected[k] = {**v}
+            else:
+                for t in sorted(tables):
+                    expected[f'{t}.{k}'] = {**v}
+        for _, v in expected.items():
+            del v['tables']
+        options_seen = set(data['no-json'])
+
+        self.schema = {}
+
+        def option_to_json_key(s):
+            return self.to_identifier(s, '', False)
+
+        # Walk through the json information building the schema as we
+        # go. This verifies consistency between command-line options
+        # and the json section of the data and builds up a schema by
+        # populating with help information as available.
+        def build_schema(j, s):
+            for k, v in j.items():
+                if not (k in expected or
+                        k.startswith('_') or
+                        isinstance(v, str)):
+                    raise Exception(f'json: unknown key {k}')
+                if k.startswith('_'):
+                    schema_key = k[1:]
+                else:
+                    schema_key = re.sub(r'[^\.]+\.', '', k)
+                schema_key = option_to_json_key(schema_key)
+                schema_value = v
+                if k in expected:
+                    options_seen.add(re.sub('^_', '', k))
+                    if v is None:
+                        schema_value = re.sub(
+                            r'--(\S+)',
+                            lambda x: option_to_json_key(x.group(1)),
+                            expected[k]['help'])
+                if (isinstance(v, dict)):
+                    schema_value = {}
+                    build_schema(v, schema_value)
+                elif (isinstance(v, list)):
+                    if len(v) != 1:
+                        raise Exception('json contains array with length != 1')
+                    if isinstance(v[0], dict):
+                        schema_value = [{}]
+                        build_schema(v[0], schema_value[0])
+                elif schema_value is None:
+                    raise Exception(f'unknown schema value for {k}')
+                s[schema_key] = schema_value
+
+        build_schema(data['json'], self.schema)
+        if options_seen != set(expected.keys()):
+            raise Exception('missing from json: ' +
+                            str(set(expected.keys()) - options_seen))
 
     def check_keys(self, what, d, exp):
         if not isinstance(d, dict):
@@ -351,7 +461,8 @@ class Main:
             exit(f'{what}: unknown keys = {extra}')
 
     def validate(self, data):
-        self.check_keys('top', data, set(['choices', 'options']))
+        self.check_keys('top', data, set(
+            ['choices', 'options', 'no-json', 'json']))
         for o in data['options']:
             self.check_keys('top', o, set(
                 ['table', 'prefix', 'bare', 'positional',
@@ -363,7 +474,10 @@ class Main:
         if const:
             identifier = f'{prefix}_{identifier.upper()}'
         else:
-            identifier = f'{prefix}_{identifier.lower()}'
+            if prefix:
+                identifier = f'{prefix}_{identifier.lower()}'
+            else:
+                identifier = identifier.lower()
             identifier = re.sub(r'_([a-z])',
                                 lambda x: x.group(1).upper(),
                                 identifier).replace('_', '')
diff --git a/job.sums b/job.sums
index 2387d64b..07c08c43 100644
--- a/job.sums
+++ b/job.sums
@@ -1,8 +1,9 @@
 # Generated by generate_auto_job
-generate_auto_job b70f64314f1ae1f100fa6a11975dee5f7669038e2a619b6c9da1e5230db1dd1b
-job.yml 8177cadf41096efdc174f04daadfe5d98c592ad44ad10cb96537521fd79a801a
+generate_auto_job 0758b244fc4e2d3e440883072d2740bc4cdb26c5aa8de938f028afd7d83fad79
+job.yml 2856c2635d42f0a58717d3ffce3125816d8f98ff17245c4b7a0669d70cd68b84
 libqpdf/qpdf/auto_job_decl.hh 97395ecbe590b23ae04d6cce2080dbd0e998917ff5eeaa5c6aafa91041d3cd6a
 libqpdf/qpdf/auto_job_help.hh 2653faaf59415bec81c3a85d426239d52b609ac24faba34ec2d26f00710dd2c6
 libqpdf/qpdf/auto_job_init.hh 465bf46769559ceb77110d1b9d3293ba9b3595850b49848c31aeabd10aadb4ad
+libqpdf/qpdf/auto_job_schema.hh c91a4e182e088797b70dda94af03ca32d360f3564890132da2a8bdc3c4432423
 manual/_ext/qpdf.py 855fe12de5af7a10bb24be6ecc4d5dff4c84ac58cf388a13be6bbb394346a67d
 manual/cli.rst b136c7f33a538c580b081a7e802c27635aad2a4229efa0eb0736466116b7aa90
diff --git a/job.yml b/job.yml
index fa15e290..dec43e90 100644
--- a/job.yml
+++ b/job.yml
@@ -217,3 +217,161 @@ options:
     required_parameter:
       prefix: prefix
       password: password
+no-json:
+  - preserve-unreferenced-resources
+json:
+  # The structure of this section defines what the json input to
+  # QPDFJob looks like. If a key starts with underscore or has a value
+  # that is a string, it does not map to a command-line argument. If
+  # value is null, its properties and help come from other information
+  # known by generate_auto_job. This information is used to construct
+  # a "schema" (as in JSON.hh) for the json input to QPDFJob. The
+  # leading underscore is removed.
+  _input:
+    _file:
+      _name: "input filename"
+      main.password:
+      password-file:
+    empty:
+  _output:
+    _file:
+      _name: "output filename"
+    replace-input:
+    split-pages:
+    _options:
+      qdf:
+      preserve-unreferenced:
+      newline-before-endstream:
+      normalize-content:
+      stream-data:
+      compress-streams:
+      recompress-flate:
+      decode-level:
+      decrypt:
+      static-aes-iv:
+      static-id:
+      no-original-object-ids:
+      copy-encryption:
+      encryption-file-password:
+      linearize:
+      linearize-pass1:
+      object-streams:
+      min-version:
+      force-version:
+      progress:
+      encrypt:
+        user-password: "user password"
+        owner-password: "owner password"
+        key-length: "key length: 48, 128, 256"
+        _40-bit:
+          Enc40.annotate:
+          Enc40.extract:
+          Enc40.modify:
+          Enc40.print:
+        _128-bit:
+          Enc128.accessibility:
+          Enc128.annotate:
+          Enc128.assemble:
+          Enc128.cleartext-metadata:
+          Enc128.extract:
+          Enc128.form:
+          Enc128.modify-other:
+          Enc128.modify:
+          Enc128.print:
+          force-V4:
+          use-aes:
+        _256-bit:
+          Enc256.accessibility:
+          Enc256.annotate:
+          Enc256.assemble:
+          Enc256.cleartext-metadata:
+          Enc256.extract:
+          Enc256.form:
+          Enc256.modify-other:
+          Enc256.modify:
+          Enc256.print:
+          allow-insecure:
+          force-R5:
+  _options:
+    allow-weak-crypto:
+    deterministic-id:
+    keep-files-open:
+    keep-files-open-threshold:
+    no-warn:
+    verbose:
+    warning-exit-0:
+    ignore-xref-streams:
+    password-is-hex-key:
+    password-mode:
+    suppress-password-recovery:
+    suppress-recovery:
+  _inspect:
+    check:
+    check-linearization:
+    filtered-stream-data:
+    is-encrypted:
+    raw-stream-data:
+    requires-password:
+    show-encryption:
+    show-encryption-key:
+    show-linearization:
+    show-npages:
+    show-object:
+    show-pages:
+    show-xref:
+    with-images:
+    list-attachments:
+    show-attachment:
+    json:
+    json-key:
+      - null
+    json-object:
+      - null
+  _transform:
+    coalesce-contents:
+    compression-level:
+    externalize-inline-images:
+    ii-min-bytes:
+    remove-unreferenced-resources:
+  _modify:
+    add-attachment:
+      - file: "attachment to add"
+        creationdate:
+        description:
+        filename:
+        key:
+        mimetype:
+        moddate:
+        replace:
+    remove-attachment:
+    copy-attachments-from:
+      - file: "attachment source filename"
+        CopyAtt.password:
+        prefix:
+    collate:
+    flatten-annotations:
+    flatten-rotation:
+    generate-appearances:
+    keep-inline-images:
+    oi-min-area:
+    oi-min-height:
+    oi-min-width:
+    optimize-images:
+    pages:
+      - file: "source for for pages"
+        Pages.password:
+        range: "page range"
+    remove-page-labels:
+    rotate:
+    overlay:
+      file: "source file for overlay"
+      UO.password:
+      from:
+      repeat:
+      to:
+    underlay:
+      file: "source file for underlay"
+      UO.password:
+      from:
+      repeat:
+      to:
diff --git a/libqpdf/qpdf/auto_job_schema.hh b/libqpdf/qpdf/auto_job_schema.hh
new file mode 100644
index 00000000..9ecdc507
--- /dev/null
+++ b/libqpdf/qpdf/auto_job_schema.hh
@@ -0,0 +1,176 @@
+static constexpr char const* JOB_SCHEMA_DATA = R"({
+  "input": {
+    "file": {
+      "name": "input filename",
+      "password": "specify password",
+      "passwordFile": "read password from a file"
+    },
+    "empty": "empty input file"
+  },
+  "output": {
+    "file": {
+      "name": "output filename"
+    },
+    "replaceInput": "replace input with output",
+    "splitPages": "write pages to separate files",
+    "options": {
+      "qdf": "enable viewing PDF code in a text editor",
+      "preserveUnreferenced": "preserve unreferenced objects",
+      "newlineBeforeEndstream": "force a newline before endstream",
+      "normalizeContent": "fix newlines in content streams",
+      "streamData": "control stream compression",
+      "compressStreams": "compress uncompressed streams",
+      "recompressFlate": "uncompress and recompress flate",
+      "decodeLevel": "control which streams to uncompress",
+      "decrypt": "remove encryption from input file",
+      "staticAesIv": "use a fixed AES vector",
+      "staticId": "use a fixed document ID",
+      "noOriginalObjectIds": "omit original object ID in qdf",
+      "copyEncryption": "copy another file's encryption details",
+      "encryptionFilePassword": "supply password for copyEncryption",
+      "linearize": "linearize (web-optimize) output",
+      "linearizePass1": "save pass 1 of linearization",
+      "objectStreams": "control use of object streams",
+      "minVersion": "set minimum PDF version",
+      "forceVersion": "set output PDF version",
+      "progress": "show progress when writing",
+      "encrypt": {
+        "userPassword": "user password",
+        "ownerPassword": "owner password",
+        "keyLength": "key length: 48, 128, 256",
+        "40Bit": {
+          "annotate": "restrict document annotation",
+          "extract": "restrict text/graphic extraction",
+          "modify": "restrict document modification",
+          "print": "restrict printing"
+        },
+        "128Bit": {
+          "accessibility": "restrict document accessibility",
+          "annotate": "restrict document annotation",
+          "assemble": "restrict document assembly",
+          "cleartextMetadata": "don't encrypt metadata",
+          "extract": "restrict text/graphic extraction",
+          "form": "restrict form filling",
+          "modifyOther": "restrict other modifications",
+          "modify": "restrict document modification",
+          "print": "restrict printing",
+          "forceV4": "force V=4 in encryption dictionary",
+          "useAes": "use AES with 128-bit encryption"
+        },
+        "256Bit": {
+          "accessibility": "restrict document accessibility",
+          "annotate": "restrict document annotation",
+          "assemble": "restrict document assembly",
+          "cleartextMetadata": "don't encrypt metadata",
+          "extract": "restrict text/graphic extraction",
+          "form": "restrict form filling",
+          "modifyOther": "restrict other modifications",
+          "modify": "restrict document modification",
+          "print": "restrict printing",
+          "allowInsecure": "allow empty owner passwords",
+          "forceR5": "use unsupported R=5 encryption"
+        }
+      }
+    }
+  },
+  "options": {
+    "allowWeakCrypto": "allow insecure cryptographic algorithms",
+    "deterministicId": "generate ID deterministically",
+    "keepFilesOpen": "manage keeping multiple files open",
+    "keepFilesOpenThreshold": "set threshold for keepFilesOpen",
+    "noWarn": "suppress printing warning messages",
+    "verbose": "print additional information",
+    "warningExit0": "exit 0 even with warnings",
+    "ignoreXrefStreams": "use xref tables rather than streams",
+    "passwordIsHexKey": "provide hex-encoded encryption key",
+    "passwordMode": "tweak how qpdf encodes passwords",
+    "suppressPasswordRecovery": "don't try different password encodings",
+    "suppressRecovery": "suppress error recovery"
+  },
+  "inspect": {
+    "check": "partially check whether PDF is valid",
+    "checkLinearization": "check linearization tables",
+    "filteredStreamData": "show filtered stream data",
+    "isEncrypted": "silently test whether a file is encrypted",
+    "rawStreamData": "show raw stream data",
+    "requiresPassword": "silently test a file's password",
+    "showEncryption": "information about encrypted files",
+    "showEncryptionKey": "show key with showEncryption",
+    "showLinearization": "show linearization hint tables",
+    "showNpages": "show number of pages",
+    "showObject": "show contents of an object",
+    "showPages": "display page dictionary information",
+    "showXref": "show cross reference data",
+    "withImages": "include image details with showPages",
+    "listAttachments": "list embedded files",
+    "showAttachment": "export an embedded file",
+    "json": "show file in json format",
+    "jsonKey": [
+      null
+    ],
+    "jsonObject": [
+      null
+    ]
+  },
+  "transform": {
+    "coalesceContents": "combine content streams",
+    "compressionLevel": "set compression level for flate",
+    "externalizeInlineImages": "convert inline to regular images",
+    "iiMinBytes": "set minimum size for externalizeInlineImages",
+    "removeUnreferencedResources": "remove unreferenced page resources"
+  },
+  "modify": {
+    "addAttachment": [
+      {
+        "file": "attachment to add",
+        "creationdate": "set attachment's creation date",
+        "description": "set attachment's description",
+        "filename": "set attachment's displayed filename",
+        "key": "specify attachment key",
+        "mimetype": "attachment mime type, e.g. application/pdf",
+        "moddate": "set attachment's modification date",
+        "replace": "replace attachment with same key"
+      }
+    ],
+    "removeAttachment": "remove an embedded file",
+    "copyAttachmentsFrom": [
+      {
+        "file": "attachment source filename",
+        "password": "specify password",
+        "prefix": "key prefix for copying attachments"
+      }
+    ],
+    "collate": "collate with pages",
+    "flattenAnnotations": "push annotations into content",
+    "flattenRotation": "remove rotation from page dictionary",
+    "generateAppearances": "generate appearances for form fields",
+    "keepInlineImages": "exclude inline images from optimization",
+    "oiMinArea": "minimum area for optimizeImages",
+    "oiMinHeight": "minimum height for optimizeImages",
+    "oiMinWidth": "minimum width for optimizeImages",
+    "optimizeImages": "use efficient compression for images",
+    "pages": [
+      {
+        "file": "source for for pages",
+        "password": "specify password",
+        "range": "page range"
+      }
+    ],
+    "removePageLabels": "remove explicit page numbers",
+    "rotate": "rotate pages",
+    "overlay": {
+      "file": "source file for overlay",
+      "password": "specify password",
+      "from": "source pages for underlay/overlay",
+      "repeat": "overlay/underlay pages to repeat",
+      "to": "destination pages for underlay/overlay"
+    },
+    "underlay": {
+      "file": "source file for underlay",
+      "password": "specify password",
+      "from": "source pages for underlay/overlay",
+      "repeat": "overlay/underlay pages to repeat",
+      "to": "destination pages for underlay/overlay"
+    }
+  }
+})";