Bump version to 3.2.0

Previously, PFERD just silently never matched the -name-> arrow. Now, it errors
when loading the config file.

CHANGELOG.md

@@ -22,6 +22,56 @@ ambiguous situations.
 
 ## Unreleased
 
+## 3.2.0 - 2021-08-04
+
+### Added
+- `--skip` command line option
+- Support for ILIAS booking objects
+
+### Changed
+- Using multiple path segments on left side of `-name->` now results in an
+  error. This was already forbidden by the documentation but silently accepted
+  by PFERD.
+- More consistent path printing in some `--explain` messages
+
+### Fixed
+- Nondeterministic name deduplication due to ILIAS reordering elements
+- More exceptions are handled properly
+
+## 3.1.0 - 2021-06-13
+
+If your config file doesn't do weird things with transforms, it should continue
+to work. If your `-re->` arrows behave weirdly, try replacing them with
+`-exact-re->` arrows. If you're on Windows, you might need to switch from `\`
+path separators to `/` in your regex rules.
+
+### Added
+- `skip` option for crawlers
+- Rules with `>>` instead of `>` as arrow head
+- `-exact-re->` arrow (behaves like `-re->` did previously)
+
+### Changed
+- The `-re->` arrow can now rename directories (like `-->`)
+- Use `/` instead of `\` as path separator for (regex) rules on Windows
+- Use the label to the left for exercises instead of the button name to
+  determine the folder name
+
+### Fixed
+- Video pagination handling in ILIAS crawler
+
+## 3.0.1 - 2021-06-01
+
+### Added
+- `credential-file` authenticator
+- `--credential-file` option for `kit-ilias-web` command
+- Warning if using concurrent tasks with `kit-ilias-web`
+
+### Changed
+- Cookies are now stored in a text-based format
+
+### Fixed
+- Date parsing now also works correctly in non-group exercises
+
 ## 3.0.0 - 2021-05-31
 
 ### Added

CONFIG.md

@@ -49,6 +49,9 @@ see the type's [documentation](#crawler-types) below. The following options are
 common to all crawlers:
 
 - `type`: The available types are specified in [this section](#crawler-types).
+- `skip`: Whether the crawler should be skipped during normal execution. The
+  crawler can still be executed manually using the `--crawler` or `-C` flags.
+  (Default: `no`)
 - `output_dir`: The directory the crawler synchronizes files to. A crawler will
   never place any files outside of this directory. (Default: the crawler's name)
 - `redownload`: When to download a file that is already present locally.
@@ -180,6 +183,22 @@ via the terminal.
 - `username`: The username. (Optional)
 - `password`: The password. (Optional)
 
+### The `credential-file` authenticator
+
+This authenticator reads a username and a password from a credential file.
+
+- `path`: Path to the credential file. (Required)
+
+The credential file has exactly two lines (trailing newline optional). The first
+line starts with `username=` and contains the username, the second line starts
+with `password=` and contains the password. The username and password may
+contain any characters except a line break.
+
+```
+username=AzureDiamond
+password=hunter2
+```
+
 ### The `keyring` authenticator
 
 This authenticator uses the system keyring to store passwords. The username can
@@ -203,56 +222,87 @@ This authenticator does not support usernames.
 Transformation rules are rules for renaming and excluding files and directories.
 They are specified line-by-line in a crawler's `transform` option. When a
 crawler needs to apply a rule to a path, it goes through this list top-to-bottom
-and choose the first matching rule.
+and applies the first matching rule.
 
 To see this process in action, you can use the `--debug-transforms` or flag or
 the `--explain` flag.
 
-Each line has the format `SOURCE ARROW TARGET` where `TARGET` is optional.
-`SOURCE` is either a normal path without spaces (e. g. `foo/bar`), or a string
-literal delimited by `"` or `'` (e. g. `"foo\" bar/baz"`). Python's string
-escape syntax is supported. Trailing slashes are ignored. `TARGET` can be
-formatted like `SOURCE`, but it can also be a single exclamation mark without
-quotes (`!`). `ARROW` is one of `-->`, `-name->`, `-exact->`, `-re->` and
-`-name-re->`
+Each rule has the format `SOURCE ARROW TARGET` (e. g. `foo/bar --> foo/baz`).
+The arrow specifies how the source and target are interpreted. The different
+kinds of arrows are documented below.
 
-If a rule's target is `!`, this means that when the rule matches on a path, the
-corresponding file or directory is ignored. If a rule's target is missing, the
-path is matched but not modified.
+`SOURCE` and `TARGET` are either a bunch of characters without spaces (e. g.
+`foo/bar`) or string literals (e. g, `"foo/b a r"`). The former syntax has no
+concept of escaping characters, so the backslash is just another character. The
+string literals however support Python's escape syntax (e. g.
+`"foo\\bar\tbaz"`). This also means that in string literals, backslashes must be
+escaped.
+
+`TARGET` can additionally be a single exclamation mark `!` (*not* `"!"`). When a
+rule with a `!` as target matches a path, the corresponding file or directory is
+ignored by the crawler instead of renamed.
+
+`TARGET` can also be omitted entirely. When a rule without target matches a
+path, the path is returned unmodified. This is useful to prevent rules further
+down from matching instead.
+
+Each arrow's behaviour can be modified slightly by changing the arrow's head
+from `>` to `>>`. When a rule with a `>>` arrow head matches a path, it doesn't
+return immediately like a normal arrow. Instead, it replaces the current path
+with its output and continues on to the next rule. In effect, this means that
+multiple rules can be applied sequentially.
 
 ### The `-->` arrow
 
-The `-->` arrow is a basic renaming operation. If a path begins with `SOURCE`,
-that part of the path is replaced with `TARGET`. This means that the rule
-`foo/bar --> baz` would convert `foo/bar` into `baz`, but also `foo/bar/xyz`
-into `baz/xyz`. The rule `foo --> !` would ignore a directory named `foo` as
-well as all its contents.
+The `-->` arrow is a basic renaming operation for files and directories. If a
+path matches `SOURCE`, it is renamed to `TARGET`.
+
+Example: `foo/bar --> baz`
+- Doesn't match `foo`, `a/foo/bar` or `foo/baz`
+- Converts `foo/bar` into `baz`
+- Converts `foo/bar/wargl` into `bar/wargl`
+
+Example: `foo/bar --> !`
+- Doesn't match `foo`, `a/foo/bar` or `foo/baz`
+- Ignores `foo/bar` and any of its children
 
 ### The `-name->` arrow
 
 The `-name->` arrow lets you rename files and directories by their name,
 regardless of where they appear in the file tree. Because of this, its `SOURCE`
 must not contain multiple path segments, only a single name. This restriction
-does not apply to its `TARGET`. The `-name->` arrow is not applied recursively
-to its own output to prevent infinite loops.
+does not apply to its `TARGET`.
 
-For example, the rule `foo -name-> bar/baz` would convert `a/foo` into
-`a/bar/baz` and `a/foo/b/c/foo` into `a/bar/baz/b/c/bar/baz`. The rule `foo
--name-> !` would ignore all directories and files named `foo`.
+Example: `foo -name-> bar/baz`
+- Doesn't match `a/foobar/b` or `x/Foo/y/z`
+- Converts `hello/foo` into `hello/bar/baz`
+- Converts `foo/world` into `bar/baz/world`
+- Converts `a/foo/b/c/foo` into `a/bar/baz/b/c/bar/baz`
+
+Example: `foo -name-> !`
+- Doesn't match `a/foobar/b` or `x/Foo/y/z`
+- Ignores any path containing a segment `foo`
 
 ### The `-exact->` arrow
 
-The `-exact->` arrow requires the path to match `SOURCE` exactly. This means
-that the rule `foo/bar -exact-> baz` would still convert `foo/bar` into `baz`,
-but `foo/bar/xyz` would be unaffected. Also, `foo -exact-> !` would only ignore
-`foo`, but not its contents (if it has any). The examples below show why this is
-useful.
+The `-exact->` arrow requires the path to match `SOURCE` exactly. The examples
+below show why this is useful.
+
+Example: `foo/bar -exact-> baz`
+- Doesn't match `foo`, `a/foo/bar` or `foo/baz`
+- Converts `foo/bar` into `baz`
+- Doesn't match `foo/bar/wargl`
+
+Example: `foo/bar -exact-> !`
+- Doesn't match `foo`, `a/foo/bar` or `foo/baz`
+- Ignores only `foo/bar`, not its children
 
 ### The `-re->` arrow
 
-The `-re->` arrow uses regular expressions. `SOURCE` is a regular expression
-that must match the entire path. If this is the case, then the capturing groups
-are available in `TARGET` for formatting.
+The `-re->` arrow is like the `-->` arrow but with regular expressions. `SOURCE`
+is a regular expression and `TARGET` an f-string based template. If a path
+matches `SOURCE`, the output path is created using `TARGET` as template.
+`SOURCE` is automatically anchored.
 
 `TARGET` uses Python's [format string syntax][3]. The *n*-th capturing group can
 be referred to as `{g<n>}` (e. g. `{g3}`). `{g0}` refers to the original path.
@@ -269,18 +319,37 @@ can use `{i3:05}`.
 PFERD even allows you to write entire expressions inside the curly braces, for
 example `{g2.lower()}` or `{g3.replace(' ', '_')}`.
 
+Example: `f(oo+)/be?ar -re-> B{g1.upper()}H/fear`
+- Doesn't match `a/foo/bar`, `foo/abc/bar`, `afoo/bar` or `foo/bars`
+- Converts `foo/bar` into `BOOH/fear`
+- Converts `fooooo/bear` into `BOOOOOH/fear`
+- Converts `foo/bar/baz` into `BOOH/fear/baz`
+
 [3]: <https://docs.python.org/3/library/string.html#format-string-syntax> "Format String Syntax"
 
 ### The `-name-re->` arrow
 
 The `-name-re>` arrow is like a combination of the `-name->` and `-re->` arrows.
-Instead of the `SOURCE` being the name of a directory or file, it's a regex that
-is matched against the names of directories and files. `TARGET` works like the
-`-re->` arrow's target.
 
-For example, the arrow `(.*)\.jpeg -name-re-> {g1}.jpg` will rename all `.jpeg`
-extensions into `.jpg`. The arrow `\..+ -name-re-> !` will ignore all files and
-directories starting with `.`.
+Example: `(.*)\.jpeg -name-re-> {g1}.jpg`
+- Doesn't match `foo/bar.png`, `baz.JPEG` or `hello,jpeg`
+- Converts `foo/bar.jpeg` into `foo/bar.jpg`
+- Converts `foo.jpeg/bar/baz.jpeg` into `foo.jpg/bar/baz.jpg`
+
+Example: `\..+ -name-re-> !`
+- Doesn't match `.`, `test`, `a.b`
+- Ignores all files and directories starting with `.`.
+
+### The `-exact-re->` arrow
+
+The `-exact-re>` arrow is like a combination of the `-exact->` and `-re->`
+arrows.
+
+Example: `f(oo+)/be?ar -exactre-> B{g1.upper()}H/fear`
+- Doesn't match `a/foo/bar`, `foo/abc/bar`, `afoo/bar` or `foo/bars`
+- Converts `foo/bar` into `BOOH/fear`
+- Converts `fooooo/bear` into `BOOOOOH/fear`
+- Doesn't match `foo/bar/baz`
 
 ### Example: Tutorials
 
@@ -307,8 +376,7 @@ tutorials --> !
 The second rule is required for many crawlers since they use the rules to decide
 which directories to crawl. If it was missing when the crawler looks at
 `tutorials/`, the third rule would match. This means the crawler would not crawl
-the `tutorials/` directory and thus not discover that `tutorials/tut02/`
-existed.
+the `tutorials/` directory and thus not discover that `tutorials/tut02/` exists.
 
 Since the second rule is only relevant for crawling, the `TARGET` is left out.
 
@@ -333,9 +401,9 @@ To do this, you can use the most powerful of arrows: The regex arrow.
 
 Note the escaped backslashes on the `SOURCE` side.
 
-### Example: Crawl a python project
+### Example: Crawl a Python project
 
-You are crawling a python project and want to ignore all hidden files (files
+You are crawling a Python project and want to ignore all hidden files (files
 whose name starts with a `.`), all `__pycache__` directories and all markdown
 files (for some weird reason).
 
@@ -355,11 +423,21 @@ README.md
 ...
 ```
 
-For this task, the name arrows can be used. They are variants of the normal
-arrows that only look at the file name instead of the entire path.
+For this task, the name arrows can be used.
 
 ```
 \..*        -name-re-> !
 __pycache__ -name->    !
 .*\.md      -name-re-> !
 ```
+
+### Example: Clean up names
+
+You want to convert all paths into lowercase and replace spaces with underscores
+before applying any rules. This can be achieved using the `>>` arrow heads.
+
+```
+(.*) -re->> "{g1.lower().replace(' ', '_')}"
+
+<other rules go here>
+```

PFERD/__main__.py

@@ -5,7 +5,8 @@ import os
 import sys
 from pathlib import Path
 
-from .cli import PARSER, load_default_section
+from .auth import AuthLoadError
+from .cli import PARSER, ParserLoadError, load_default_section
 from .config import Config, ConfigDumpError, ConfigLoadError, ConfigOptionError
 from .logging import log
 from .pferd import Pferd, PferdLoadError
@@ -36,6 +37,9 @@ def load_config(args: argparse.Namespace) -> Config:
         log.error(str(e))
         log.error_contd(e.reason)
         sys.exit(1)
+    except ParserLoadError as e:
+        log.error(str(e))
+        sys.exit(1)
 
 
 def configure_logging_from_args(args: argparse.Namespace) -> None:
@@ -112,7 +116,7 @@ def main() -> None:
         sys.exit()
 
     try:
-        pferd = Pferd(config, args.crawler)
+        pferd = Pferd(config, args.crawler, args.skip)
     except PferdLoadError as e:
         log.unlock()
         log.error(str(e))
@@ -131,7 +135,7 @@ def main() -> None:
             loop.close()
         else:
             asyncio.run(pferd.run(args.debug_transforms))
-    except ConfigOptionError as e:
+    except (ConfigOptionError, AuthLoadError) as e:
         log.unlock()
         log.error(str(e))
         sys.exit(1)
@@ -143,7 +147,6 @@ def main() -> None:
         log.unlock()
         log.explain_topic("Interrupted, exiting immediately")
         log.explain("Open files and connections are left for the OS to clean up")
-        log.explain("Temporary files are not cleaned up")
         pferd.print_report()
         # TODO Clean up tmp files
         # And when those files *do* actually get cleaned up properly,

PFERD/auth/__init__.py

@@ -2,7 +2,8 @@ from configparser import SectionProxy
 from typing import Callable, Dict
 
 from ..config import Config
-from .authenticator import Authenticator, AuthError, AuthSection  # noqa: F401
+from .authenticator import Authenticator, AuthError, AuthLoadError, AuthSection  # noqa: F401
+from .credential_file import CredentialFileAuthenticator, CredentialFileAuthSection
 from .keyring import KeyringAuthenticator, KeyringAuthSection
 from .simple import SimpleAuthenticator, SimpleAuthSection
 from .tfa import TfaAuthenticator
@@ -14,10 +15,12 @@ AuthConstructor = Callable[[
 ], Authenticator]
 
 AUTHENTICATORS: Dict[str, AuthConstructor] = {
+    "credential-file": lambda n, s, c:
+        CredentialFileAuthenticator(n, CredentialFileAuthSection(s), c),
+    "keyring": lambda n, s, c:
+        KeyringAuthenticator(n, KeyringAuthSection(s)),
     "simple": lambda n, s, c:
         SimpleAuthenticator(n, SimpleAuthSection(s)),
     "tfa": lambda n, s, c:
         TfaAuthenticator(n),
-    "keyring": lambda n, s, c:
-        KeyringAuthenticator(n, KeyringAuthSection(s))
 }

PFERD/auth/authenticator.py

@@ -13,14 +13,15 @@ class AuthError(Exception):
 
 
 class AuthSection(Section):
-    pass
+    def type(self) -> str:
+        value = self.s.get("type")
+        if value is None:
+            self.missing_value("type")
+        return value
 
 
 class Authenticator(ABC):
-    def __init__(
-            self,
-            name: str
-    ) -> None:
+    def __init__(self, name: str) -> None:
         """
         Initialize an authenticator from its name and its section in the config
         file.

PFERD/auth/credential_file.py

@@ -0,0 +1,44 @@
+from pathlib import Path
+from typing import Tuple
+
+from ..config import Config
+from ..utils import fmt_real_path
+from .authenticator import Authenticator, AuthLoadError, AuthSection
+
+
+class CredentialFileAuthSection(AuthSection):
+    def path(self) -> Path:
+        value = self.s.get("path")
+        if value is None:
+            self.missing_value("path")
+        return Path(value)
+
+
+class CredentialFileAuthenticator(Authenticator):
+    def __init__(self, name: str, section: CredentialFileAuthSection, config: Config) -> None:
+        super().__init__(name)
+
+        path = config.default_section.working_dir() / section.path()
+        try:
+            with open(path) as f:
+                lines = list(f)
+        except OSError as e:
+            raise AuthLoadError(f"No credential file at {fmt_real_path(path)}") from e
+
+        if len(lines) != 2:
+            raise AuthLoadError("Credential file must be two lines long")
+        [uline, pline] = lines
+        uline = uline[:-1]  # Remove trailing newline
+        if pline.endswith("\n"):
+            pline = pline[:-1]
+
+        if not uline.startswith("username="):
+            raise AuthLoadError("First line must start with 'username='")
+        if not pline.startswith("password="):
+            raise AuthLoadError("Second line must start with 'password='")
+
+        self._username = uline[9:]
+        self._password = pline[9:]
+
+    async def credentials(self) -> Tuple[str, str]:
+        return self._username, self._password

PFERD/auth/keyring.py

@@ -18,11 +18,7 @@ class KeyringAuthSection(AuthSection):
 
 class KeyringAuthenticator(Authenticator):
 
-    def __init__(
-            self,
-            name: str,
-            section: KeyringAuthSection,
-    ) -> None:
+    def __init__(self, name: str, section: KeyringAuthSection) -> None:
         super().__init__(name)
 
         self._username = section.username()

PFERD/auth/simple.py

@@ -14,11 +14,7 @@ class SimpleAuthSection(AuthSection):
 
 
 class SimpleAuthenticator(Authenticator):
-    def __init__(
-            self,
-            name: str,
-            section: SimpleAuthSection,
-    ) -> None:
+    def __init__(self, name: str, section: SimpleAuthSection) -> None:
         super().__init__(name)
 
         self._username = section.username()

PFERD/auth/tfa.py

@@ -6,10 +6,7 @@ from .authenticator import Authenticator, AuthError
 
 
 class TfaAuthenticator(Authenticator):
-    def __init__(
-            self,
-            name: str,
-    ) -> None:
+    def __init__(self, name: str) -> None:
         super().__init__(name)
 
     async def username(self) -> str:

PFERD/cli/__init__.py

@@ -1,11 +1,12 @@
 # isort: skip_file
 
 # The order of imports matters because each command module registers itself
-# with the parser from ".parser". Because of this, isort is disabled for this
+# with the parser from ".parser" and the import order affects the order in
+# which they appear in the help. Because of this, isort is disabled for this
 # file. Also, since we're reexporting or just using the side effect of
 # importing itself, we get a few linting warnings, which we're disabling as
 # well.
 
 from . import command_local  # noqa: F401 imported but unused
 from . import command_kit_ilias_web  # noqa: F401 imported but unused
-from .parser import PARSER, load_default_section  # noqa: F401 imported but unused
+from .parser import PARSER, ParserLoadError, load_default_section  # noqa: F401 imported but unused

PFERD/cli/command_kit_ilias_web.py

@@ -4,7 +4,8 @@ from pathlib import Path
 
 from ..crawl.ilias.file_templates import Links
 from ..logging import log
-from .parser import CRAWLER_PARSER, SUBPARSERS, BooleanOptionalAction, load_crawler, show_value_error
+from .parser import (CRAWLER_PARSER, SUBPARSERS, BooleanOptionalAction, ParserLoadError, load_crawler,
+                     show_value_error)
 
 SUBPARSER = SUBPARSERS.add_parser(
     "kit-ilias-web",
@@ -38,6 +39,12 @@ GROUP.add_argument(
     action=BooleanOptionalAction,
     help="use the system keyring to store and retrieve passwords"
 )
+GROUP.add_argument(
+    "--credential-file",
+    type=Path,
+    metavar="PATH",
+    help="read username and password from a credential file"
+)
 GROUP.add_argument(
     "--links",
     type=show_value_error(Links.from_string),
@@ -88,11 +95,19 @@ def load(
 
     parser["auth:ilias"] = {}
     auth_section = parser["auth:ilias"]
+    if args.credential_file is not None:
+        if args.username is not None:
+            raise ParserLoadError("--credential-file and --username can't be used together")
+        if args.keyring:
+            raise ParserLoadError("--credential-file and --keyring can't be used together")
+        auth_section["type"] = "credential-file"
+        auth_section["path"] = str(args.credential_file)
+    elif args.keyring:
+        auth_section["type"] = "keyring"
+    else:
         auth_section["type"] = "simple"
     if args.username is not None:
         auth_section["username"] = args.username
-    if args.keyring:
-        auth_section["type"] = "keyring"
 
 
 SUBPARSER.set_defaults(command=load)

PFERD/cli/parser.py

@@ -8,6 +8,10 @@ from ..output_dir import OnConflict, Redownload
 from ..version import NAME, VERSION
 
 
+class ParserLoadError(Exception):
+    pass
+
+
 # TODO Replace with argparse version when updating to 3.9?
 class BooleanOptionalAction(argparse.Action):
     def __init__(
@@ -177,6 +181,14 @@ PARSER.add_argument(
     help="only execute a single crawler."
     " Can be specified multiple times to execute multiple crawlers"
 )
+PARSER.add_argument(
+    "--skip", "-S",
+    action="append",
+    type=str,
+    metavar="NAME",
+    help="don't execute this particular crawler."
+    " Can be specified multiple times to skip multiple crawlers"
+)
 PARSER.add_argument(
     "--working-dir",
     type=Path,

PFERD/config.py

@@ -69,6 +69,7 @@ class Section:
 
 class DefaultSection(Section):
     def working_dir(self) -> Path:
+        # TODO Change to working dir instead of manually prepending it to paths
         pathstr = self.s.get("working_dir", ".")
         return Path(pathstr).expanduser()
 

PFERD/crawl/__init__.py

@@ -3,7 +3,7 @@ from typing import Callable, Dict
 
 from ..auth import Authenticator
 from ..config import Config
-from .crawler import Crawler, CrawlError  # noqa: F401
+from .crawler import Crawler, CrawlError, CrawlerSection  # noqa: F401
 from .ilias import KitIliasWebCrawler, KitIliasWebCrawlerSection
 from .local_crawler import LocalCrawler, LocalCrawlerSection
 

PFERD/crawl/crawler.py

@@ -56,7 +56,7 @@ def noncritical(f: Wrapped) -> Wrapped:
     return wrapper  # type: ignore
 
 
-AWrapped = TypeVar("AWrapped", bound=Callable[..., Awaitable[None]])
+AWrapped = TypeVar("AWrapped", bound=Callable[..., Awaitable[Optional[Any]]])
 
 
 def anoncritical(f: AWrapped) -> AWrapped:
@@ -72,14 +72,14 @@ def anoncritical(f: AWrapped) -> AWrapped:
     Warning: Must only be applied to member functions of the Crawler class!
     """
 
-    async def wrapper(*args: Any, **kwargs: Any) -> None:
+    async def wrapper(*args: Any, **kwargs: Any) -> Optional[Any]:
         if not (args and isinstance(args[0], Crawler)):
             raise RuntimeError("@anoncritical must only applied to Crawler methods")
 
         crawler = args[0]
 
         try:
-            await f(*args, **kwargs)
+            return await f(*args, **kwargs)
         except (CrawlWarning, OutputDirError, MarkDuplicateError, MarkConflictError) as e:
             log.warn(str(e))
             crawler.error_free = False
@@ -87,6 +87,8 @@ def anoncritical(f: AWrapped) -> AWrapped:
             crawler.error_free = False
             raise
 
+        return None
+
     return wrapper  # type: ignore
 
 
@@ -132,6 +134,15 @@ class DownloadToken(ReusableAsyncContextManager[Tuple[ProgressBar, FileSink]]):
 
 
 class CrawlerSection(Section):
+    def type(self) -> str:
+        value = self.s.get("type")
+        if value is None:
+            self.missing_value("type")
+        return value
+
+    def skip(self) -> bool:
+        return self.s.getboolean("skip", fallback=False)
+
     def output_dir(self, name: str) -> Path:
         # TODO Use removeprefix() after switching to 3.9
         if name.startswith("crawl:"):
@@ -309,6 +320,7 @@ class Crawler(ABC):
             log.explain("Warnings or errors occurred during this run")
             log.explain("Answer: No")
 
+    @anoncritical
     async def run(self) -> None:
         """
         Start the crawling process. Call this function if you want to use a

PFERD/crawl/http_crawler.py

@@ -1,7 +1,8 @@
 import asyncio
+import http.cookies
 import ssl
 from pathlib import Path, PurePath
-from typing import Dict, List, Optional
+from typing import Any, Dict, List, Optional
 
 import aiohttp
 import certifi
@@ -105,6 +106,25 @@ class HttpCrawler(Crawler):
 
         self._shared_cookie_jar_paths.append(self._cookie_jar_path)
 
+    def _load_cookies_from_file(self, path: Path) -> None:
+        jar: Any = http.cookies.SimpleCookie()
+        with open(path) as f:
+            for i, line in enumerate(f):
+                # Names of headers are case insensitive
+                if line[:11].lower() == "set-cookie:":
+                    jar.load(line[11:])
+                else:
+                    log.explain(f"Line {i} doesn't start with 'Set-Cookie:', ignoring it")
+        self._cookie_jar.update_cookies(jar)
+
+    def _save_cookies_to_file(self, path: Path) -> None:
+        jar: Any = http.cookies.SimpleCookie()
+        for morsel in self._cookie_jar:
+            jar[morsel.key] = morsel
+        with open(path, "w") as f:
+            f.write(jar.output(sep="\n"))
+            f.write("\n")  # A trailing newline is just common courtesy
+
     def _load_cookies(self) -> None:
         log.explain_topic("Loading cookies")
 
@@ -134,7 +154,7 @@ class HttpCrawler(Crawler):
 
         log.explain(f"Loading cookies from {fmt_real_path(cookie_jar_path)}")
         try:
-            self._cookie_jar.load(cookie_jar_path)
+            self._load_cookies_from_file(cookie_jar_path)
         except Exception as e:
             log.explain("Failed to load cookies")
             log.explain(str(e))
@@ -144,7 +164,7 @@ class HttpCrawler(Crawler):
 
         try:
             log.explain(f"Saving cookies to {fmt_real_path(self._cookie_jar_path)}")
-            self._cookie_jar.save(self._cookie_jar_path)
+            self._save_cookies_to_file(self._cookie_jar_path)
         except Exception as e:
             log.warn(f"Failed to save cookies to {fmt_real_path(self._cookie_jar_path)}")
             log.warn(str(e))

PFERD/crawl/ilias/kit_ilias_html.py

@@ -22,6 +22,7 @@ class IliasElementType(Enum):
     FOLDER = "folder"
     FORUM = "forum"
     LINK = "link"
+    BOOKING = "booking"
     MEETING = "meeting"
     VIDEO = "video"
     VIDEO_PLAYER = "video_player"
@@ -37,6 +38,17 @@ class IliasPageElement:
     mtime: Optional[datetime] = None
     description: Optional[str] = None
 
+    def id(self) -> str:
+        regexes = [r"eid=(?P<id>[0-9a-z\-]+)", r"file_(?P<id>\d+)", r"ref_id=(?P<id>\d+)"]
+
+        for regex in regexes:
+            if match := re.search(regex, self.url):
+                return match.groupdict()["id"]
+
+        # Fall back to URL
+        log.warn(f"Didn't find identity for {self.name} - {self.url}. Please report this.")
+        return self.url
+
 
 class IliasPage:
 
@@ -62,9 +74,11 @@ class IliasPage:
         log.explain("Page is a normal folder, searching for elements")
         return self._find_normal_entries()
 
-    def get_next_stage_url(self) -> Optional[str]:
+    def get_next_stage_element(self) -> Optional[IliasPageElement]:
         if self._is_ilias_opencast_embedding():
-            return self.get_child_elements()[0].url
+            return self.get_child_elements()[0]
+        if self._page_type == IliasElementType.VIDEO_FOLDER_MAYBE_PAGINATED:
+            return self._find_video_entries_paginated()[0]
         return None
 
     def _is_video_player(self) -> bool:
@@ -230,12 +244,16 @@ class IliasPage:
             parent_row: Tag = link.findParent("tr")
             children: List[Tag] = parent_row.findChildren("td")
 
-            # <checkbox> <name> <uploader> <date> <download>
-            #     0         1        2       3        4
             name = _sanitize_path_name(children[1].getText().strip())
-            date = demangle_date(children[3].getText().strip())
-
             log.explain(f"Found exercise detail entry {name!r}")
+
+            for child in reversed(children):
+                date = demangle_date(child.getText().strip(), fail_silently=True)
+                if date is not None:
+                    break
+            if date is None:
+                log.warn(f"Date parsing failed for exercise entry {name!r}")
+
             results.append(IliasPageElement(
                 IliasElementType.FILE,
                 self._abs_url_from_link(link),
@@ -289,7 +307,13 @@ class IliasPage:
 
             # Add each listing as a new
             for listing in file_listings:
-                file_name = _sanitize_path_name(listing.getText().strip())
+                parent_container: Tag = listing.findParent(
+                    "div", attrs={"class": lambda x: x and "form-group" in x}
+                )
+                label_container: Tag = parent_container.find(
+                    attrs={"class": lambda x: x and "control-label" in x}
+                )
+                file_name = _sanitize_path_name(label_container.getText().strip())
                 url = self._abs_url_from_link(listing)
                 log.explain(f"Found exercise detail {file_name!r} at {url}")
                 results.append(IliasPageElement(
@@ -470,7 +494,7 @@ class IliasPage:
             return None
 
         if "opencast" in str(img_tag["alt"]).lower():
-            return IliasElementType.VIDEO_FOLDER
+            return IliasElementType.VIDEO_FOLDER_MAYBE_PAGINATED
 
         if str(img_tag["src"]).endswith("icon_exc.svg"):
             return IliasElementType.EXERCISE
@@ -478,6 +502,9 @@ class IliasPage:
         if str(img_tag["src"]).endswith("icon_webr.svg"):
             return IliasElementType.LINK
 
+        if str(img_tag["src"]).endswith("icon_book.svg"):
+            return IliasElementType.BOOKING
+
         if str(img_tag["src"]).endswith("frm.svg"):
             return IliasElementType.FORUM
 
@@ -522,7 +549,7 @@ german_months = ['Jan', 'Feb', 'Mär', 'Apr', 'Mai', 'Jun', 'Jul', 'Aug', 'Sep',
 english_months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
 
 
-def demangle_date(date_str: str) -> Optional[datetime]:
+def demangle_date(date_str: str, fail_silently: bool = False) -> Optional[datetime]:
     """
     Demangle a given date in one of the following formats:
     "Gestern, HH:MM"
@@ -554,6 +581,7 @@ def demangle_date(date_str: str) -> Optional[datetime]:
 
         return datetime(year, month, day, hour, minute)
     except Exception:
+        if not fail_silently:
             log.warn(f"Date parsing failed for {date_str!r}")
         return None
 

PFERD/crawl/ilias/kit_ilias_web_crawler.py

@@ -12,7 +12,7 @@ from ...config import Config
 from ...logging import ProgressBar, log
 from ...output_dir import FileSink, Redownload
 from ...utils import fmt_path, soupify, url_set_query_param
-from ..crawler import CrawlError, CrawlWarning, anoncritical
+from ..crawler import CrawlError, CrawlToken, CrawlWarning, DownloadToken, anoncritical
 from ..http_crawler import HttpCrawler, HttpCrawlerSection
 from .file_templates import Links
 from .kit_ilias_html import IliasElementType, IliasPage, IliasPageElement
@@ -21,7 +21,6 @@ TargetType = Union[str, int]
 
 
 class KitIliasWebCrawlerSection(HttpCrawlerSection):
-
     def target(self) -> TargetType:
         target = self.s.get("target")
         if not target:
@@ -82,17 +81,16 @@ _VIDEO_ELEMENTS: Set[IliasElementType] = set([
     IliasElementType.VIDEO_FOLDER_MAYBE_PAGINATED,
 ])
 
-AWrapped = TypeVar("AWrapped", bound=Callable[..., Awaitable[None]])
+AWrapped = TypeVar("AWrapped", bound=Callable[..., Awaitable[Optional[Any]]])
 
 
 def _iorepeat(attempts: int, name: str) -> Callable[[AWrapped], AWrapped]:
     def decorator(f: AWrapped) -> AWrapped:
-        async def wrapper(*args: Any, **kwargs: Any) -> None:
+        async def wrapper(*args: Any, **kwargs: Any) -> Optional[Any]:
             last_exception: Optional[BaseException] = None
             for round in range(attempts):
                 try:
-                    await f(*args, **kwargs)
-                    return
+                    return await f(*args, **kwargs)
                 except aiohttp.ContentTypeError:  # invalid content type
                     raise CrawlWarning("ILIAS returned an invalid content type")
                 except aiohttp.TooManyRedirects:
@@ -164,6 +162,12 @@ class KitIliasWebCrawler(HttpCrawler):
         auth = section.auth(authenticators)
         super().__init__(name, section, config, shared_auth=auth)
 
+        if section.tasks() > 1:
+            log.warn("""
+Please avoid using too many parallel requests as these are the KIT ILIAS
+instance's greatest bottleneck.
+            """.strip())
+
         self._shibboleth_login = KitShibbolethLogin(
             auth,
             section.tfa_auth(authenticators),
@@ -225,17 +229,34 @@ class KitIliasWebCrawler(HttpCrawler):
 
         # Fill up our task list with the found elements
         await gather_elements()
-        tasks = [self._handle_ilias_element(PurePath("."), element) for element in elements]
+
+        elements.sort(key=lambda e: e.id())
+
+        tasks: List[Awaitable[None]] = []
+        for element in elements:
+            if handle := await self._handle_ilias_element(PurePath("."), element):
+                tasks.append(asyncio.create_task(handle))
 
         # And execute them
         await self.gather(tasks)
 
-    async def _handle_ilias_page(self, url: str, parent: IliasPageElement, path: PurePath) -> None:
+    async def _handle_ilias_page(
+        self,
+        url: str,
+        parent: IliasPageElement,
+        path: PurePath,
+    ) -> Optional[Awaitable[None]]:
         maybe_cl = await self.crawl(path)
         if not maybe_cl:
-            return
-        cl = maybe_cl  # Not mypy's fault, but explained here: https://github.com/python/mypy/issues/2608
+            return None
+        return self._crawl_ilias_page(url, parent, maybe_cl)
 
+    async def _crawl_ilias_page(
+        self,
+        url: str,
+        parent: IliasPageElement,
+        cl: CrawlToken,
+    ) -> None:
         elements: List[IliasPageElement] = []
 
         @_iorepeat(3, "crawling folder")
@@ -243,19 +264,30 @@ class KitIliasWebCrawler(HttpCrawler):
             elements.clear()
             async with cl:
                 next_stage_url: Optional[str] = url
+                current_parent = parent
 
                 while next_stage_url:
                     soup = await self._get_page(next_stage_url)
-                    log.explain_topic(f"Parsing HTML page for {fmt_path(path)}")
+                    log.explain_topic(f"Parsing HTML page for {fmt_path(cl.path)}")
                     log.explain(f"URL: {next_stage_url}")
-                    page = IliasPage(soup, url, parent)
-                    next_stage_url = page.get_next_stage_url()
+                    page = IliasPage(soup, next_stage_url, current_parent)
+                    if next_element := page.get_next_stage_element():
+                        current_parent = next_element
+                        next_stage_url = next_element.url
+                    else:
+                        next_stage_url = None
 
                 elements.extend(page.get_child_elements())
 
         # Fill up our task list with the found elements
         await gather_elements()
-        tasks = [self._handle_ilias_element(cl.path, element) for element in elements]
+
+        elements.sort(key=lambda e: e.id())
+
+        tasks: List[Awaitable[None]] = []
+        for element in elements:
+            if handle := await self._handle_ilias_element(cl.path, element):
+                tasks.append(asyncio.create_task(handle))
 
         # And execute them
         await self.gather(tasks)
@@ -264,7 +296,11 @@ class KitIliasWebCrawler(HttpCrawler):
     # Shouldn't happen but we also really don't want to let I/O errors bubble up to anoncritical.
     # If that happens we will be terminated as anoncritical doesn't tream them as non-critical.
     @_wrap_io_in_warning("handling ilias element")
-    async def _handle_ilias_element(self, parent_path: PurePath, element: IliasPageElement) -> None:
+    async def _handle_ilias_element(
+        self,
+        parent_path: PurePath,
+        element: IliasPageElement,
+    ) -> Optional[Awaitable[None]]:
         element_path = PurePath(parent_path, element.name)
 
         if element.type in _VIDEO_ELEMENTS:
@@ -272,35 +308,43 @@ class KitIliasWebCrawler(HttpCrawler):
             if not self._videos:
                 log.explain("Video crawling is disabled")
                 log.explain("Answer: no")
-                return
+                return None
             else:
                 log.explain("Video crawling is enabled")
                 log.explain("Answer: yes")
 
         if element.type == IliasElementType.FILE:
-            await self._download_file(element, element_path)
+            return await self._handle_file(element, element_path)
         elif element.type == IliasElementType.FORUM:
             log.explain_topic(f"Decision: Crawl {fmt_path(element_path)}")
             log.explain("Forums are not supported")
             log.explain("Answer: No")
+            return None
         elif element.type == IliasElementType.TEST:
             log.explain_topic(f"Decision: Crawl {fmt_path(element_path)}")
             log.explain("Tests contain no relevant files")
             log.explain("Answer: No")
+            return None
         elif element.type == IliasElementType.LINK:
-            await self._download_link(element, element_path)
+            return await self._handle_link(element, element_path)
+        elif element.type == IliasElementType.BOOKING:
+            return await self._handle_booking(element, element_path)
         elif element.type == IliasElementType.VIDEO:
-            await self._download_file(element, element_path)
+            return await self._handle_file(element, element_path)
         elif element.type == IliasElementType.VIDEO_PLAYER:
-            await self._download_video(element, element_path)
+            return await self._handle_video(element, element_path)
         elif element.type in _DIRECTORY_PAGES:
-            await self._handle_ilias_page(element.url, element, element_path)
+            return await self._handle_ilias_page(element.url, element, element_path)
         else:
             # This will retry it a few times, failing everytime. It doesn't make any network
             # requests, so that's fine.
             raise CrawlWarning(f"Unknown element type: {element.type!r}")
 
-    async def _download_link(self, element: IliasPageElement, element_path: PurePath) -> None:
+    async def _handle_link(
+        self,
+        element: IliasPageElement,
+        element_path: PurePath,
+    ) -> Optional[Awaitable[None]]:
         log.explain_topic(f"Decision: Crawl Link {fmt_path(element_path)}")
         log.explain(f"Links type is {self._links}")
 
@@ -308,32 +352,72 @@ class KitIliasWebCrawler(HttpCrawler):
         link_extension = self._links.extension()
         if not link_template_maybe or not link_extension:
             log.explain("Answer: No")
-            return
+            return None
         else:
             log.explain("Answer: Yes")
-        link_template = link_template_maybe
         element_path = element_path.with_name(element_path.name + link_extension)
 
         maybe_dl = await self.download(element_path, mtime=element.mtime)
         if not maybe_dl:
-            return
-        dl = maybe_dl  # Not mypy's fault, but explained here: https://github.com/python/mypy/issues/2608
+            return None
+
+        return self._download_link(element, link_template_maybe, maybe_dl)
 
     @_iorepeat(3, "resolving link")
-        async def impl() -> None:
+    async def _download_link(self, element: IliasPageElement, link_template: str, dl: DownloadToken) -> None:
         async with dl as (bar, sink):
             export_url = element.url.replace("cmd=calldirectlink", "cmd=exportHTML")
             real_url = await self._resolve_link_target(export_url)
+            self._write_link_content(link_template, real_url, element.name, element.description, sink)
 
+    def _write_link_content(
+        self,
+        link_template: str,
+        url: str,
+        name: str,
+        description: Optional[str],
+        sink: FileSink,
+    ) -> None:
         content = link_template
-                content = content.replace("{{link}}", real_url)
-                content = content.replace("{{name}}", element.name)
-                content = content.replace("{{description}}", str(element.description))
+        content = content.replace("{{link}}", url)
+        content = content.replace("{{name}}", name)
+        content = content.replace("{{description}}", str(description))
         content = content.replace("{{redirect_delay}}", str(self._link_file_redirect_delay))
         sink.file.write(content.encode("utf-8"))
         sink.done()
 
-        await impl()
+    async def _handle_booking(
+        self,
+        element: IliasPageElement,
+        element_path: PurePath,
+    ) -> Optional[Awaitable[None]]:
+        log.explain_topic(f"Decision: Crawl Booking Link {fmt_path(element_path)}")
+        log.explain(f"Links type is {self._links}")
+
+        link_template_maybe = self._links.template()
+        link_extension = self._links.extension()
+        if not link_template_maybe or not link_extension:
+            log.explain("Answer: No")
+            return None
+        else:
+            log.explain("Answer: Yes")
+        element_path = element_path.with_name(element_path.name + link_extension)
+
+        maybe_dl = await self.download(element_path, mtime=element.mtime)
+        if not maybe_dl:
+            return None
+
+        return self._download_booking(element, link_template_maybe, maybe_dl)
+
+    @_iorepeat(3, "resolving booking")
+    async def _download_booking(
+        self,
+        element: IliasPageElement,
+        link_template: str,
+        dl: DownloadToken,
+    ) -> None:
+        async with dl as (bar, sink):
+            self._write_link_content(link_template, element.url, element.name, element.description, sink)
 
     async def _resolve_link_target(self, export_url: str) -> str:
         async with self.session.get(export_url, allow_redirects=False) as resp:
@@ -350,16 +434,20 @@ class KitIliasWebCrawler(HttpCrawler):
 
         raise CrawlError("resolve_link_target failed even after authenticating")
 
-    async def _download_video(self, element: IliasPageElement, element_path: PurePath) -> None:
+    async def _handle_video(
+        self,
+        element: IliasPageElement,
+        element_path: PurePath,
+    ) -> Optional[Awaitable[None]]:
         # Videos will NOT be redownloaded - their content doesn't really change and they are chunky
         maybe_dl = await self.download(element_path, mtime=element.mtime, redownload=Redownload.NEVER)
         if not maybe_dl:
-            return
-        dl = maybe_dl  # Not mypy's fault, but explained here: https://github.com/python/mypy/issues/2608
+            return None
+
+        return self._download_video(element, maybe_dl)
 
     @_iorepeat(3, "downloading video")
-        async def impl() -> None:
-            assert dl  # The function is only reached when dl is not None
+    async def _download_video(self, element: IliasPageElement, dl: DownloadToken) -> None:
         async with dl as (bar, sink):
             page = IliasPage(await self._get_page(element.url), element.url, element)
             real_element = page.get_child_elements()[0]
@@ -368,22 +456,22 @@ class KitIliasWebCrawler(HttpCrawler):
 
             await self._stream_from_url(real_element.url, sink, bar, is_video=True)
 
-        await impl()
-
-    async def _download_file(self, element: IliasPageElement, element_path: PurePath) -> None:
+    async def _handle_file(
+        self,
+        element: IliasPageElement,
+        element_path: PurePath,
+    ) -> Optional[Awaitable[None]]:
         maybe_dl = await self.download(element_path, mtime=element.mtime)
         if not maybe_dl:
-            return
-        dl = maybe_dl  # Not mypy's fault, but explained here: https://github.com/python/mypy/issues/2608
+            return None
+        return self._download_file(element, maybe_dl)
 
     @_iorepeat(3, "downloading file")
-        async def impl() -> None:
+    async def _download_file(self, element: IliasPageElement, dl: DownloadToken) -> None:
         assert dl  # The function is only reached when dl is not None
         async with dl as (bar, sink):
             await self._stream_from_url(element.url, sink, bar, is_video=False)
 
-        await impl()
-
     async def _stream_from_url(self, url: str, sink: FileSink, bar: ProgressBar, is_video: bool) -> None:
         async def try_stream() -> bool:
             async with self.session.get(url, allow_redirects=is_video) as resp:

PFERD/pferd.py

@@ -3,9 +3,9 @@ from typing import Dict, List, Optional
 
 from rich.markup import escape
 
-from .auth import AUTHENTICATORS, Authenticator, AuthError
+from .auth import AUTHENTICATORS, Authenticator, AuthError, AuthSection
 from .config import Config, ConfigOptionError
-from .crawl import CRAWLERS, Crawler, CrawlError, KitIliasWebCrawler
+from .crawl import CRAWLERS, Crawler, CrawlError, CrawlerSection, KitIliasWebCrawler
 from .logging import log
 from .utils import fmt_path
 
@@ -15,30 +15,33 @@ class PferdLoadError(Exception):
 
 
 class Pferd:
-    def __init__(self, config: Config, cli_crawlers: Optional[List[str]]):
+    def __init__(self, config: Config, cli_crawlers: Optional[List[str]], cli_skips: Optional[List[str]]):
         """
         May throw PferdLoadError.
         """
 
         self._config = config
-        self._crawlers_to_run = self._find_crawlers_to_run(config, cli_crawlers)
+        self._crawlers_to_run = self._find_crawlers_to_run(config, cli_crawlers, cli_skips)
 
         self._authenticators: Dict[str, Authenticator] = {}
         self._crawlers: Dict[str, Crawler] = {}
 
-    def _find_crawlers_to_run(self, config: Config, cli_crawlers: Optional[List[str]]) -> List[str]:
-        log.explain_topic("Deciding which crawlers to run")
-        crawl_sections = [name for name, _ in config.crawl_sections()]
+    def _find_config_crawlers(self, config: Config) -> List[str]:
+        crawl_sections = []
+
+        for name, section in config.crawl_sections():
+            if CrawlerSection(section).skip():
+                log.explain(f"Skipping {name!r}")
+            else:
+                crawl_sections.append(name)
 
-        if cli_crawlers is None:
-            log.explain("No crawlers specified on CLI")
-            log.explain("Running all crawlers specified in config")
         return crawl_sections
 
+    def _find_cli_crawlers(self, config: Config, cli_crawlers: List[str]) -> List[str]:
         if len(cli_crawlers) != len(set(cli_crawlers)):
             raise PferdLoadError("Some crawlers were selected multiple times")
 
-        log.explain("Crawlers specified on CLI")
+        crawl_sections = [name for name, _ in config.crawl_sections()]
 
         crawlers_to_run = []  # With crawl: prefix
         unknown_names = []  # Without crawl: prefix
@@ -62,10 +65,36 @@ class Pferd:
 
         return crawlers_to_run
 
+    def _find_crawlers_to_run(
+            self,
+            config: Config,
+            cli_crawlers: Optional[List[str]],
+            cli_skips: Optional[List[str]],
+    ) -> List[str]:
+        log.explain_topic("Deciding which crawlers to run")
+
+        crawlers: List[str]
+        if cli_crawlers is None:
+            log.explain("No crawlers specified on CLI")
+            log.explain("Running crawlers specified in config")
+            crawlers = self._find_config_crawlers(config)
+        else:
+            log.explain("Crawlers specified on CLI")
+            crawlers = self._find_cli_crawlers(config, cli_crawlers)
+
+        skips = {f"crawl:{name}" for name in cli_skips} if cli_skips else set()
+        for crawler in crawlers:
+            if crawler in skips:
+                log.explain(f"Skipping crawler {crawler!r}")
+        crawlers = [crawler for crawler in crawlers if crawler not in skips]
+
+        return crawlers
+
     def _load_authenticators(self) -> None:
         for name, section in self._config.auth_sections():
             log.print(f"[bold bright_cyan]Loading[/] {escape(name)}")
-            auth_type = section.get("type")
+
+            auth_type = AuthSection(section).type()
             authenticator_constructor = AUTHENTICATORS.get(auth_type)
             if authenticator_constructor is None:
                 raise ConfigOptionError(name, "type", f"Unknown authenticator type: {auth_type!r}")
@@ -80,7 +109,7 @@ class Pferd:
         for name, section in self._config.crawl_sections():
             log.print(f"[bold bright_cyan]Loading[/] {escape(name)}")
 
-            crawl_type = section.get("type")
+            crawl_type = CrawlerSection(section).type()
             crawler_constructor = CRAWLERS.get(crawl_type)
             if crawler_constructor is None:
                 raise ConfigOptionError(name, "type", f"Unknown crawler type: {crawl_type!r}")

PFERD/transformer.py

@@ -1,151 +1,166 @@
-# I'm sorry that this code has become a bit dense and unreadable. While
-# reading, it is important to remember what True and False mean. I'd love to
-# have some proper sum-types for the inputs and outputs, they'd make this code
-# a lot easier to understand.
-
 import ast
 import re
 from abc import ABC, abstractmethod
+from dataclasses import dataclass
+from enum import Enum
 from pathlib import PurePath
-from typing import Dict, Optional, Sequence, Union
+from typing import Callable, Dict, List, Optional, Sequence, TypeVar, Union
 
 from .logging import log
-from .utils import fmt_path
+from .utils import fmt_path, str_path
 
 
-class Rule(ABC):
-    @abstractmethod
-    def transform(self, path: PurePath) -> Union[PurePath, bool]:
-        """
-        Try to apply this rule to the path. Returns another path if the rule
-        was successfully applied, True if the rule matched but resulted in an
-        exclamation mark, and False if the rule didn't match at all.
-        """
+class ArrowHead(Enum):
+    NORMAL = 0
+    SEQUENCE = 1
 
+
+class Ignore:
     pass
 
 
-# These rules all use a Union[T, bool] for their right side. They are passed a
-# T if the arrow's right side was a normal string, True if it was an
-# exclamation mark and False if it was missing entirely.
+class Empty:
+    pass
 
-class NormalRule(Rule):
-    def __init__(self, left: PurePath, right: Union[PurePath, bool]):
 
-        self._left = left
-        self._right = right
+RightSide = Union[str, Ignore, Empty]
 
-    def _match_prefix(self, path: PurePath) -> Optional[PurePath]:
-        left_parts = list(reversed(self._left.parts))
-        path_parts = list(reversed(path.parts))
 
-        if len(left_parts) > len(path_parts):
+@dataclass
+class Transformed:
+    path: PurePath
+
+
+class Ignored:
+    pass
+
+
+TransformResult = Optional[Union[Transformed, Ignored]]
+
+
+@dataclass
+class Rule:
+    left: str
+    left_index: int
+    name: str
+    head: ArrowHead
+    right: RightSide
+    right_index: int
+
+    def right_result(self, path: PurePath) -> Union[str, Transformed, Ignored]:
+        if isinstance(self.right, str):
+            return self.right
+        elif isinstance(self.right, Ignore):
+            return Ignored()
+        elif isinstance(self.right, Empty):
+            return Transformed(path)
+        else:
+            raise RuntimeError(f"Right side has invalid type {type(self.right)}")
+
+
+class Transformation(ABC):
+    def __init__(self, rule: Rule):
+        self.rule = rule
+
+    @abstractmethod
+    def transform(self, path: PurePath) -> TransformResult:
+        pass
+
+
+class ExactTf(Transformation):
+    def transform(self, path: PurePath) -> TransformResult:
+        if path != PurePath(self.rule.left):
             return None
 
-        while left_parts and path_parts:
-            left_part = left_parts.pop()
-            path_part = path_parts.pop()
+        right = self.rule.right_result(path)
+        if not isinstance(right, str):
+            return right
 
-            if left_part != path_part:
+        return Transformed(PurePath(right))
+
+
+class ExactReTf(Transformation):
+    def transform(self, path: PurePath) -> TransformResult:
+        match = re.fullmatch(self.rule.left, str_path(path))
+        if not match:
             return None
 
-        if left_parts:
-            return None
+        right = self.rule.right_result(path)
+        if not isinstance(right, str):
+            return right
 
-        path_parts.reverse()
-        return PurePath(*path_parts)
-
-    def transform(self, path: PurePath) -> Union[PurePath, bool]:
-        if rest := self._match_prefix(path):
-            if isinstance(self._right, bool):
-                return self._right or path
-            else:
-                return self._right / rest
-
-        return False
-
-
-class ExactRule(Rule):
-    def __init__(self, left: PurePath, right: Union[PurePath, bool]):
-        self._left = left
-        self._right = right
-
-    def transform(self, path: PurePath) -> Union[PurePath, bool]:
-        if path == self._left:
-            if isinstance(self._right, bool):
-                return self._right or path
-            else:
-                return self._right
-
-        return False
-
-
-class NameRule(Rule):
-    def __init__(self, subrule: Rule):
-        self._subrule = subrule
-
-    def transform(self, path: PurePath) -> Union[PurePath, bool]:
-        matched = False
-        result = PurePath()
-
-        for part in path.parts:
-            part_result = self._subrule.transform(PurePath(part))
-            if isinstance(part_result, PurePath):
-                matched = True
-                result /= part_result
-            elif part_result:
-                # If any subrule call ignores its path segment, the entire path
-                # should be ignored
-                return True
-            else:
-                # The subrule doesn't modify this segment, but maybe other
-                # segments
-                result /= part
-
-        if matched:
-            return result
-        else:
-            # The subrule has modified no segments, so this name version of it
-            # doesn't match
-            return False
-
-
-class ReRule(Rule):
-    def __init__(self, left: str, right: Union[str, bool]):
-        self._left = left
-        self._right = right
-
-    def transform(self, path: PurePath) -> Union[PurePath, bool]:
-        if match := re.fullmatch(self._left, str(path)):
-            if isinstance(self._right, bool):
-                return self._right or path
-
-            vars: Dict[str, Union[str, int, float]] = {}
-
-            # For some reason, mypy thinks that "groups" has type List[str].
-            # But since elements of "match.groups()" can be None, mypy is
-            # wrong.
+        # For some reason, mypy thinks that "groups" has type List[str]. But
+        # since elements of "match.groups()" can be None, mypy is wrong.
         groups: Sequence[Optional[str]] = [match[0]] + list(match.groups())
+
+        locals_dir: Dict[str, Union[str, int, float]] = {}
         for i, group in enumerate(groups):
             if group is None:
                 continue
 
-                vars[f"g{i}"] = group
+            locals_dir[f"g{i}"] = group
 
             try:
-                    vars[f"i{i}"] = int(group)
+                locals_dir[f"i{i}"] = int(group)
             except ValueError:
                 pass
 
             try:
-                    vars[f"f{i}"] = float(group)
+                locals_dir[f"f{i}"] = float(group)
             except ValueError:
                 pass
 
-            result = eval(f"f{self._right!r}", vars)
-            return PurePath(result)
+        result = eval(f"f{right!r}", {}, locals_dir)
+        return Transformed(PurePath(result))
 
-        return False
+
+class RenamingParentsTf(Transformation):
+    def __init__(self, sub_tf: Transformation):
+        super().__init__(sub_tf.rule)
+        self.sub_tf = sub_tf
+
+    def transform(self, path: PurePath) -> TransformResult:
+        for i in range(len(path.parts), -1, -1):
+            parent = PurePath(*path.parts[:i])
+            child = PurePath(*path.parts[i:])
+
+            transformed = self.sub_tf.transform(parent)
+            if not transformed:
+                continue
+            elif isinstance(transformed, Transformed):
+                return Transformed(transformed.path / child)
+            elif isinstance(transformed, Ignored):
+                return transformed
+            else:
+                raise RuntimeError(f"Invalid transform result of type {type(transformed)}: {transformed}")
+
+        return None
+
+
+class RenamingPartsTf(Transformation):
+    def __init__(self, sub_tf: Transformation):
+        super().__init__(sub_tf.rule)
+        self.sub_tf = sub_tf
+
+    def transform(self, path: PurePath) -> TransformResult:
+        result = PurePath()
+        any_part_matched = False
+        for part in path.parts:
+            transformed = self.sub_tf.transform(PurePath(part))
+            if not transformed:
+                result /= part
+            elif isinstance(transformed, Transformed):
+                result /= transformed.path
+                any_part_matched = True
+            elif isinstance(transformed, Ignored):
+                return transformed
+            else:
+                raise RuntimeError(f"Invalid transform result of type {type(transformed)}: {transformed}")
+
+        if any_part_matched:
+            return Transformed(result)
+        else:
+            return None
 
 
 class RuleParseError(Exception):
@@ -162,18 +177,15 @@ class RuleParseError(Exception):
         log.error_contd(f"{spaces}^--- {self.reason}")
 
 
+T = TypeVar("T")
+
+
 class Line:
     def __init__(self, line: str, line_nr: int):
         self._line = line
         self._line_nr = line_nr
         self._index = 0
 
-    def get(self) -> Optional[str]:
-        if self._index < len(self._line):
-            return self._line[self._index]
-
-        return None
-
     @property
     def line(self) -> str:
         return self._line
@@ -190,155 +202,196 @@ class Line:
     def index(self, index: int) -> None:
         self._index = index
 
-    def advance(self) -> None:
-        self._index += 1
+    @property
+    def rest(self) -> str:
+        return self.line[self.index:]
 
-    def expect(self, string: str) -> None:
-        for char in string:
-            if self.get() == char:
-                self.advance()
+    def peek(self, amount: int = 1) -> str:
+        return self.rest[:amount]
+
+    def take(self, amount: int = 1) -> str:
+        string = self.peek(amount)
+        self.index += len(string)
+        return string
+
+    def expect(self, string: str) -> str:
+        if self.peek(len(string)) == string:
+            return self.take(len(string))
         else:
-                raise RuleParseError(self, f"Expected {char!r}")
+            raise RuleParseError(self, f"Expected {string!r}")
+
+    def expect_with(self, string: str, value: T) -> T:
+        self.expect(string)
+        return value
+
+    def one_of(self, parsers: List[Callable[[], T]], description: str) -> T:
+        for parser in parsers:
+            index = self.index
+            try:
+                return parser()
+            except RuleParseError:
+                self.index = index
+
+        raise RuleParseError(self, description)
+
+
+# RULE = LEFT SPACE '-' NAME '-' HEAD (SPACE RIGHT)?
+# SPACE = ' '+
+# NAME = '' | 'exact' | 'name' | 're' | 'exact-re' | 'name-re'
+# HEAD = '>' | '>>'
+# LEFT = STR | QUOTED_STR
+# RIGHT = STR | QUOTED_STR | '!'
+
+
+def parse_zero_or_more_spaces(line: Line) -> None:
+    while line.peek() == " ":
+        line.take()
+
+
+def parse_one_or_more_spaces(line: Line) -> None:
+    line.expect(" ")
+    parse_zero_or_more_spaces(line)
+
+
+def parse_str(line: Line) -> str:
+    result = []
+    while c := line.peek():
+        if c == " ":
+            break
+        else:
+            line.take()
+            result.append(c)
+
+    if result:
+        return "".join(result)
+    else:
+        raise RuleParseError(line, "Expected non-space character")
 
 
 QUOTATION_MARKS = {'"', "'"}
 
 
-def parse_string_literal(line: Line) -> str:
+def parse_quoted_str(line: Line) -> str:
     escaped = False
 
     # Points to first character of string literal
     start_index = line.index
 
-    quotation_mark = line.get()
+    quotation_mark = line.peek()
     if quotation_mark not in QUOTATION_MARKS:
-        # This should never happen as long as this function is only called from
-        # parse_string.
-        raise RuleParseError(line, "Invalid quotation mark")
-    line.advance()
+        raise RuleParseError(line, "Expected quotation mark")
+    line.take()
 
-    while c := line.get():
+    while c := line.peek():
         if escaped:
             escaped = False
-            line.advance()
+            line.take()
         elif c == quotation_mark:
-            line.advance()
+            line.take()
             stop_index = line.index
             literal = line.line[start_index:stop_index]
+            try:
                 return ast.literal_eval(literal)
+            except SyntaxError as e:
+                line.index = start_index
+                raise RuleParseError(line, str(e)) from e
         elif c == "\\":
             escaped = True
-            line.advance()
+            line.take()
         else:
-            line.advance()
+            line.take()
 
     raise RuleParseError(line, "Expected end of string literal")
 
 
-def parse_until_space_or_eol(line: Line) -> str:
-    result = []
-    while c := line.get():
-        if c == " ":
-            break
-        result.append(c)
-        line.advance()
-
-    return "".join(result)
-
-
-def parse_string(line: Line) -> Union[str, bool]:
-    if line.get() in QUOTATION_MARKS:
-        return parse_string_literal(line)
+def parse_left(line: Line) -> str:
+    if line.peek() in QUOTATION_MARKS:
+        return parse_quoted_str(line)
     else:
-        string = parse_until_space_or_eol(line)
+        return parse_str(line)
+
+
+def parse_right(line: Line) -> Union[str, Ignore]:
+    c = line.peek()
+    if c in QUOTATION_MARKS:
+        return parse_quoted_str(line)
+    else:
+        string = parse_str(line)
         if string == "!":
-            return True
+            return Ignore()
         return string
 
 
-def parse_arrow(line: Line) -> str:
-    line.expect("-")
-
-    name = []
-    while True:
-        c = line.get()
-        if not c:
-            raise RuleParseError(line, "Expected rest of arrow")
-        elif c == "-":
-            line.advance()
-            c = line.get()
-            if not c:
-                raise RuleParseError(line, "Expected rest of arrow")
-            elif c == ">":
-                line.advance()
-                break  # End of arrow
-            else:
-                name.append("-")
-                continue
-        else:
-            name.append(c)
-
-        line.advance()
-
-    return "".join(name)
+def parse_arrow_name(line: Line) -> str:
+    return line.one_of([
+        lambda: line.expect("exact-re"),
+        lambda: line.expect("exact"),
+        lambda: line.expect("name-re"),
+        lambda: line.expect("name"),
+        lambda: line.expect("re"),
+        lambda: line.expect(""),
+    ], "Expected arrow name")
 
 
-def parse_whitespace(line: Line) -> None:
-    line.expect(" ")
-    while line.get() == " ":
-        line.advance()
+def parse_arrow_head(line: Line) -> ArrowHead:
+    return line.one_of([
+        lambda: line.expect_with(">>", ArrowHead.SEQUENCE),
+        lambda: line.expect_with(">", ArrowHead.NORMAL),
+    ], "Expected arrow head")
 
 
 def parse_eol(line: Line) -> None:
-    if line.get() is not None:
+    if line.peek():
         raise RuleParseError(line, "Expected end of line")
 
 
 def parse_rule(line: Line) -> Rule:
-    # Parse left side
-    leftindex = line.index
-    left = parse_string(line)
-    if isinstance(left, bool):
-        line.index = leftindex
-        raise RuleParseError(line, "Left side can't be '!'")
-    leftpath = PurePath(left)
+    parse_zero_or_more_spaces(line)
+    left_index = line.index
+    left = parse_left(line)
 
-    # Parse arrow
-    parse_whitespace(line)
-    arrowindex = line.index
-    arrowname = parse_arrow(line)
+    parse_one_or_more_spaces(line)
 
-    # Parse right side
-    if line.get():
-        parse_whitespace(line)
-        right = parse_string(line)
-    else:
-        right = False
-    rightpath: Union[PurePath, bool]
-    if isinstance(right, bool):
-        rightpath = right
-    else:
-        rightpath = PurePath(right)
+    line.expect("-")
+    name = parse_arrow_name(line)
+    line.expect("-")
+    head = parse_arrow_head(line)
 
+    right_index = line.index
+    right: RightSide
+    try:
+        parse_zero_or_more_spaces(line)
+        parse_eol(line)
+        right = Empty()
+    except RuleParseError:
+        line.index = right_index
+        parse_one_or_more_spaces(line)
+        right = parse_right(line)
         parse_eol(line)
 
-    # Dispatch
-    if arrowname == "":
-        return NormalRule(leftpath, rightpath)
-    elif arrowname == "name":
-        if len(leftpath.parts) > 1:
-            line.index = leftindex
-            raise RuleParseError(line, "SOURCE must be a single name, not multiple segments")
-        return NameRule(ExactRule(leftpath, rightpath))
-    elif arrowname == "exact":
-        return ExactRule(leftpath, rightpath)
-    elif arrowname == "re":
-        return ReRule(left, right)
-    elif arrowname == "name-re":
-        return NameRule(ReRule(left, right))
+    return Rule(left, left_index, name, head, right, right_index)
+
+
+def parse_transformation(line: Line) -> Transformation:
+    rule = parse_rule(line)
+
+    if rule.name == "":
+        return RenamingParentsTf(ExactTf(rule))
+    elif rule.name == "exact":
+        return ExactTf(rule)
+    elif rule.name == "name":
+        if len(PurePath(rule.left).parts) > 1:
+            line.index = rule.left_index
+            raise RuleParseError(line, "Expected name, not multiple segments")
+        return RenamingPartsTf(ExactTf(rule))
+    elif rule.name == "re":
+        return RenamingParentsTf(ExactReTf(rule))
+    elif rule.name == "exact-re":
+        return ExactReTf(rule)
+    elif rule.name == "name-re":
+        return RenamingPartsTf(ExactReTf(rule))
     else:
-        line.index = arrowindex + 1  # For nicer error message
-        raise RuleParseError(line, f"Invalid arrow name {arrowname!r}")
+        raise RuntimeError(f"Invalid arrow name {rule.name!r}")
 
 
 class Transformer:
@@ -347,32 +400,40 @@ class Transformer:
         May throw a RuleParseException.
         """
 
-        self._rules = []
+        self._tfs = []
         for i, line in enumerate(rules.split("\n")):
             line = line.strip()
             if line:
-                rule = parse_rule(Line(line, i))
-                self._rules.append((line, rule))
+                tf = parse_transformation(Line(line, i))
+                self._tfs.append((line, tf))
 
     def transform(self, path: PurePath) -> Optional[PurePath]:
-        for i, (line, rule) in enumerate(self._rules):
+        for i, (line, tf) in enumerate(self._tfs):
             log.explain(f"Testing rule {i+1}: {line}")
 
             try:
-                result = rule.transform(path)
+                result = tf.transform(path)
             except Exception as e:
                 log.warn(f"Error while testing rule {i+1}: {line}")
                 log.warn_contd(str(e))
                 continue
 
-            if isinstance(result, PurePath):
-                log.explain(f"Match found, transformed path to {fmt_path(result)}")
-                return result
-            elif result:  # Exclamation mark
-                log.explain("Match found, path ignored")
-                return None
-            else:
+            if not result:
                 continue
 
-        log.explain("No rule matched, path is unchanged")
+            if isinstance(result, Ignored):
+                log.explain("Match found, path ignored")
+                return None
+
+            if tf.rule.head == ArrowHead.NORMAL:
+                log.explain(f"Match found, transformed path to {fmt_path(result.path)}")
+                path = result.path
+                break
+            elif tf.rule.head == ArrowHead.SEQUENCE:
+                log.explain(f"Match found, updated path to {fmt_path(result.path)}")
+                path = result.path
+            else:
+                raise RuntimeError(f"Invalid transform result of type {type(result)}: {result}")
+
+        log.explain(f"Final result: {fmt_path(path)}")
         return path

PFERD/utils.py

@@ -91,8 +91,14 @@ def url_set_query_params(url: str, params: Dict[str, str]) -> str:
     return result
 
 
+def str_path(path: PurePath) -> str:
+    if not path.parts:
+        return "."
+    return "/".join(path.parts)
+
+
 def fmt_path(path: PurePath) -> str:
-    return repr(str(path))
+    return repr(str_path(path))
 
 
 def fmt_real_path(path: Path) -> str:

PFERD/version.py

@@ -1,2 +1,2 @@
 NAME = "PFERD"
-VERSION = "3.0.0"
+VERSION = "3.2.0"

README.md

@@ -28,9 +28,9 @@ The use of [venv](https://docs.python.org/3/library/venv.html) is recommended.
 
 ## Basic usage
 
-PFERD can be run directly from the command line with no config file.
-Run `pferd -h` to get an overview of available commands and options.
-Run `pferd <command> -h` to see which options a command has.
+PFERD can be run directly from the command line with no config file. Run `pferd
+-h` to get an overview of available commands and options. Run `pferd <command>
+-h` to see which options a command has.
 
 For example, you can download your personal desktop from the KIT ILIAS like
 this:
@@ -116,17 +116,18 @@ transform =
   Online-Tests --> !
   Vorlesungswerbung --> !
 
+  # Rename folders
+  Lehrbücher --> Vorlesung
+  # Note the ">>" arrow head which lets us apply further rules to files moved to "Übung"
+  Übungsunterlagen -->> Übung
+
   # Move exercises to own folder. Rename them to "Blatt-XX.pdf" to make them sort properly
-  "Übungsunterlagen/(\d+). Übungsblatt.pdf" -re-> Blätter/Blatt-{i1:02}.pdf
+  "Übung/(\d+). Übungsblatt.pdf" -re-> Blätter/Blatt-{i1:02}.pdf
   # Move solutions to own folder. Rename them to "Blatt-XX-Lösung.pdf" to make them sort properly
-  "Übungsunterlagen/(\d+). Übungsblatt.*Musterlösung.pdf" -re-> Blätter/Blatt-{i1:02}-Lösung.pdf
+  "Übung/(\d+). Übungsblatt.*Musterlösung.pdf" -re-> Blätter/Blatt-{i1:02}-Lösung.pdf
 
   # The course has nested folders with the same name - flatten them
-  "Übungsunterlagen/(.+?)/\\1/(.*)" -re-> Übung/{g1}/{g2}
-
-  # Rename remaining folders
-  Übungsunterlagen --> Übung
-  Lehrbücher --> Vorlesung
+  "Übung/(.+?)/\\1" -re-> Übung/{g1}
 
 [crawl:Bar]
 type = kit-ilias-web

scripts/setup

@@ -12,6 +12,6 @@ pip install --upgrade setuptools
 # Installing PFERD itself
 pip install --editable .
 
-# Installing various tools
-pip install --upgrade mypy flake8 autopep8 isort
-pip install --upgrade pyinstaller
+# Installing tools and type hints
+pip install --upgrade mypy flake8 autopep8 isort pyinstaller
+pip install --upgrade types-chardet types-certifi