Merge branch 'eigenmagic:main' into main

CHANGELOG.md

@@ -6,7 +6,23 @@ This project uses [Semantic Versioning] and generally follows the conventions of
 
 ## [Unreleased]
 
-- Planning to add allowlist thresholds as noted in #28
+## [v0.4.3] - 2023-02-13
+
+### Added
+
+- Added Mastodon public API parser type because #33 (9fe9342)
+- Added ability to set scheme when talking to instances (9fe9342)
+- Added tests of comment merging. (fb3a7ec)
+- Added blocklist thresholds. (bb1d89e)
+- Added logging to help debug threshold-based merging. (b67ff0c)
+- Added extra documentation on configuring thresholds. (6c72af8)
+- Updated documentation to reflect Mastodon v4.1.0 changes to the application scopes screen. (b92dd21)
+
+### Changed
+
+- Dropped minimum Python version to 3.6 (df3c16f)
+- Don't merge comments if new comment is empty. (b8aa11e)
+- Tweaked comment merging to pass tests. (fb3a7ec)
 
 ## [v0.4.2] - 2023-01-19
 

README.md

@@ -41,6 +41,8 @@ appropriate.
 
  - Provides (hopefully) sensible defaults to minimise first-time setup.
  - Global and fine-grained configuration options available for those complex situations that crop up sometimes.
+ - Allowlists to override blocks in blocklists to ensure you never block instances you want to keep.
+ - Blocklist thresholds if you want to only block when an instance shows up in multiple blocklists.
 
 ## Installing
 
@@ -79,17 +81,16 @@ admin to add a new Application at
 `https://<instance-domain>/settings/applications/` and then tell you the access
 token.
 
-The application needs the `admin:read:domain_blocks` OAuth scope, but
+The application needs the `admin:read:domain_blocks` OAuth scope. You can allow
-unfortunately this scope isn't available in the current application screen
+full `admin:read` access, but be aware that this authorizes someone to read all
-(v4.0.2 of Mastodon at time of writing, but this has been fixed in the main
+the data in the instance. That's asking a lot of a remote instance admin who
-branch). 
+just wants to share domain_blocks with you.
 
-You can allow full `admin:read` access, but be aware that this authorizes
+The `admin:read:domain_blocks` scope is available as of Mastodon v4.1.0, but for
-someone to read all the data in the instance. That's asking a lot of a remote
+earlier versions admins will need to use the manual method described below.
-instance admin who just wants to share domain_blocks with you.
 
-For now, you can ask the instance admin to update the scope in the database
+You can update the scope for your application in the database directly like
-directly like this:
+this:
 
 ```
 UPDATE oauth_applications as app
@@ -134,8 +135,12 @@ chmod o-r <configfile>
 ```
 
 You can also grant full `admin:write` scope to the application, but if you'd
-prefer to keep things more tightly secured you'll need to use SQL to set the
+prefer to keep things more tightly secured, limit the scope to
-scopes in the database and then regenerate the token:
+`admin:read:domain_blocks`.
+
+Again, this scope is only available in the application config screen as of
+Mastodon v4.1.0. If your instance is on an earlier version, you'll need to use
+SQL to set the scopes in the database and then regenerate the token:
 
 ```
 UPDATE oauth_applications as app
@@ -192,6 +197,7 @@ Supported formats are currently:
 
  - Comma-Separated Values (CSV)
  - JSON
+ - Mastodon v4.1 flavoured CSV
  - RapidBlock CSV
  - RapidBlock JSON
 
@@ -209,6 +215,17 @@ A CSV format blocklist must contain a header row with at least a `domain` and `s
 
 Optional fields, as listed about, may also be included.
 
+#### Mastodon v4.1 CSV format
+
+As of v4.1.0, Mastodon can export domain blocks as a CSV file. However, in their
+infinite wisdom, the Mastodon devs decided that field names should begin with a
+`#` character in the header, unlike the field names in the JSON output via the
+API… or in pretty much any other CSV file anywhere else.
+
+Setting the format to `mastodon_csv` will strip off the `#` character when
+parsing and FediBlockHole can then use Mastodon v4.1 CSV blocklists like any
+other CSV formatted blocklist.
+
 #### JSON format
 
 JSON is also supported. It uses the same format as the JSON returned from the Mastodon API.

etc/sample.fediblockhole.conf.toml

@@ -56,6 +56,24 @@ blocklist_instance_destinations = [
 # The 'min' mergeplan will use the lightest severity block found for a domain.
 # mergeplan = 'max'
 
+## Optional threshold-based merging.
+# Only merge in domain blocks if the domain is mentioned in
+# at least `threshold` blocklists.
+# `merge_thresold` is an integer, with a default value of 0.
+# The `merge_threshold_type` can be `count` or `pct`.
+# If `count` type is selected, the threshold is reached when the domain
+# is mentioned in at least `merge_threshold` blocklists. The default value
+# of 0 means that every block in every list will be merged in.
+# If `pct` type is selected, `merge_threshold` is interpreted as a percentage,
+# i.e. if `merge_threshold` = 20, blocks will only be merged in if the domain
+# is present in at least 20% of blocklists.
+# Percentage calculated as number_of_mentions / total_number_of_blocklists.
+# The percentage method is more flexibile, but also more complicated, so take care
+# when using it.
+# 
+# merge_threshold_type = 'count'
+# merge_threshold = 0
+
 ## Set which fields we import
 ## 'domain' and 'severity' are always imported, these are additional
 ## 

pyproject.toml

@@ -1,10 +1,10 @@
 [project]
 name = "fediblockhole"
-version = "0.4.2"
+version = "0.4.3"
 description = "Federated blocklist management for Mastodon"
 readme = "README.md"
 license = {file = "LICENSE"}
-requires-python = ">=3.10"
+requires-python = ">=3.6"
 keywords = ["mastodon", "fediblock"]
 authors = [ 
     {name = "Justin Warren"}, {email = "justin@eigenmagic.com"}
@@ -17,6 +17,10 @@ classifiers = [
     "Natural Language :: English",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.7",
+    "Programming Language :: Python :: 3.6",
 ]
 dependencies = [
     "requests",

samples/demo-allowlist-01.csv

@@ -1,3 +1,4 @@
 "domain","severity","private_comment","public_comment","reject_media","reject_reports","obfuscate"
-"eigenmagic.net","noop","Never block me","Only the domain field matters",False,False,False
+"eigenmagic.net","noop","Never block me","Only the domain field matters for allowlists",False,False,False
-"example.org","noop","Never block me either","The severity is ignored as are all other fields",False,False,False
+"example.org","noop","Never block me either","The severity is ignored in allowlists as are all other fields",False,False,False
+"demo01.example.org","noop","Never block me either","But you can use them to leave yourself or others notes on why the item is here",False,False,False

src/fediblockhole/__init__.py

@@ -11,7 +11,7 @@ import os.path
 import sys
 import urllib.request as urlr
 
-from .blocklist_parser import parse_blocklist
+from .blocklists import Blocklist, parse_blocklist
 from .const import DomainBlock, BlockSeverity
 
 from importlib.metadata import version
@@ -59,19 +59,19 @@ def sync_blocklists(conf: argparse.Namespace):
     # Add extra export fields if defined in config
     export_fields.extend(conf.export_fields)
 
-    blocklists = {}
+    blocklists = []
     # Fetch blocklists from URLs
     if not conf.no_fetch_url:
-        blocklists = fetch_from_urls(blocklists, conf.blocklist_url_sources,
+        blocklists.extend(fetch_from_urls(conf.blocklist_url_sources,
-            import_fields, conf.save_intermediate, conf.savedir, export_fields)
+            import_fields, conf.save_intermediate, conf.savedir, export_fields))
 
     # Fetch blocklists from remote instances
     if not conf.no_fetch_instance:
-        blocklists = fetch_from_instances(blocklists, conf.blocklist_instance_sources,
+        blocklists.extend(fetch_from_instances(conf.blocklist_instance_sources,
-            import_fields, conf.save_intermediate, conf.savedir, export_fields)
+            import_fields, conf.save_intermediate, conf.savedir, export_fields))
 
     # Merge blocklists into an update dict
-    merged = merge_blocklists(blocklists, conf.mergeplan)
+    merged = merge_blocklists(blocklists, conf.mergeplan, conf.merge_threshold, conf.merge_threshold_type)
 
     # Remove items listed in allowlists, if any
     allowlists = fetch_allowlists(conf)
@@ -80,48 +80,48 @@ def sync_blocklists(conf: argparse.Namespace):
     # Save the final mergelist, if requested
     if conf.blocklist_savefile:
         log.info(f"Saving merged blocklist to {conf.blocklist_savefile}")
-        save_blocklist_to_file(merged.values(), conf.blocklist_savefile, export_fields)
+        save_blocklist_to_file(merged, conf.blocklist_savefile, export_fields)
 
     # Push the blocklist to destination instances
     if not conf.no_push_instance:
         log.info("Pushing domain blocks to instances...")
         for dest in conf.blocklist_instance_destinations:
-            domain = dest['domain']
+            target = dest['domain']
             token = dest['token']
             scheme = dest.get('scheme', 'https')
             max_followed_severity = BlockSeverity(dest.get('max_followed_severity', 'silence'))
-            push_blocklist(token, domain, merged.values(), conf.dryrun, import_fields, max_followed_severity, scheme)
+            push_blocklist(token, target, merged, conf.dryrun, import_fields, max_followed_severity, scheme)
 
-def apply_allowlists(merged: dict, conf: argparse.Namespace, allowlists: dict):
+def apply_allowlists(merged: Blocklist, conf: argparse.Namespace, allowlists: dict):
     """Apply allowlists
     """
     # Apply allows specified on the commandline
     for domain in conf.allow_domains:
         log.info(f"'{domain}' allowed by commandline, removing any blocks...")
-        if domain in merged:
+        if domain in merged.blocks:
-            del merged[domain]
+            del merged.blocks[domain]
 
     # Apply allows from URLs lists
     log.info("Removing domains from URL allowlists...")
-    for key, alist in allowlists.items():
+    for alist in allowlists:
-        log.debug(f"Processing allows from '{key}'...")
+        log.debug(f"Processing allows from '{alist.origin}'...")
-        for allowed in alist:
+        for allowed in alist.blocks.values():
             domain = allowed.domain
             log.debug(f"Removing allowlisted domain '{domain}' from merged list.")
-            if domain in merged:
+            if domain in merged.blocks:
-                del merged[domain]
+                del merged.blocks[domain]
 
     return merged
 
-def fetch_allowlists(conf: argparse.Namespace) -> dict:
+def fetch_allowlists(conf: argparse.Namespace) -> Blocklist:
     """
     """
     if conf.allowlist_url_sources:
-        allowlists = fetch_from_urls({}, conf.allowlist_url_sources, ALLOWLIST_IMPORT_FIELDS)
+        allowlists = fetch_from_urls(conf.allowlist_url_sources, ALLOWLIST_IMPORT_FIELDS, conf.save_intermediate, conf.savedir)
         return allowlists
-    return {}
+    return Blocklist()
 
-def fetch_from_urls(blocklists: dict, url_sources: dict,
+def fetch_from_urls(url_sources: dict,
     import_fields: list=IMPORT_FIELDS,
     save_intermediate: bool=False,
     savedir: str=None, export_fields: list=EXPORT_FIELDS) -> dict:
@@ -131,7 +131,7 @@ def fetch_from_urls(blocklists: dict, url_sources: dict,
     @returns: A dict of blocklists, same as input, but (possibly) modified
     """
     log.info("Fetching domain blocks from URLs...")
-
+    blocklists = []
     for item in url_sources:
         url = item['url']
         # If import fields are provided, they override the global ones passed in
@@ -144,14 +144,14 @@ def fetch_from_urls(blocklists: dict, url_sources: dict,
         listformat = item.get('format', 'csv')
         with urlr.urlopen(url) as fp:
             rawdata = fp.read(URL_BLOCKLIST_MAXSIZE).decode('utf-8')
-            blocklists[url] = parse_blocklist(rawdata, listformat, import_fields, max_severity)
+            bl = parse_blocklist(rawdata, url, listformat, import_fields, max_severity)
-            
+            blocklists.append(bl)
             if save_intermediate:
-            save_intermediate_blocklist(blocklists[url], url, savedir, export_fields)
+                save_intermediate_blocklist(bl, savedir, export_fields)
     
     return blocklists
 
-def fetch_from_instances(blocklists: dict, sources: dict,
+def fetch_from_instances(sources: dict,
     import_fields: list=IMPORT_FIELDS,
     save_intermediate: bool=False,
     savedir: str=None, export_fields: list=EXPORT_FIELDS) -> dict:
@@ -161,12 +161,13 @@ def fetch_from_instances(blocklists: dict, sources: dict,
     @returns: A dict of blocklists, same as input, but (possibly) modified
     """
     log.info("Fetching domain blocks from instances...")
+    blocklists = []
     for item in sources:
         domain = item['domain']
         admin = item.get('admin', False)
         token = item.get('token', None)
         scheme = item.get('scheme', 'https')
-        itemsrc = f"{scheme}://{domain}/api"
+        # itemsrc = f"{scheme}://{domain}/api"
 
         # If import fields are provided, they override the global ones passed in
         source_import_fields = item.get('import_fields', None)
@@ -174,45 +175,69 @@ def fetch_from_instances(blocklists: dict, sources: dict,
             # Ensure we always use the default fields
             import_fields = IMPORT_FIELDS.extend(source_import_fields)
 
-        # Add the blocklist with the domain as the source key
+        bl = fetch_instance_blocklist(domain, token, admin, import_fields, scheme)
-        blocklists[itemsrc] = fetch_instance_blocklist(domain, token, admin, import_fields, scheme)
+        blocklists.append(bl)
         if save_intermediate:
-            save_intermediate_blocklist(blocklists[itemsrc], domain, savedir, export_fields)
+            save_intermediate_blocklist(bl, savedir, export_fields)
     return blocklists
 
-def merge_blocklists(blocklists: dict, mergeplan: str='max') -> dict:
+def merge_blocklists(blocklists: list[Blocklist], mergeplan: str='max',
+    threshold: int=0,
+    threshold_type: str='count') -> Blocklist:
     """Merge fetched remote blocklists into a bulk update
     @param blocklists: A dict of lists of DomainBlocks, keyed by source.
         Each value is a list of DomainBlocks
     @param mergeplan: An optional method of merging overlapping block definitions
         'max' (the default) uses the highest severity block found
         'min' uses the lowest severity block found
+    @param threshold: An integer used in the threshold mechanism.
+        If a domain is not present in this number/pct or more of the blocklists,
+        it will not get merged into the final list.
+    @param threshold_type: choice of ['count', 'pct']
+        If `count`, threshold is met if block is present in `threshold`
+        or more blocklists.
+        If `pct`, theshold is met if block is present in
+        count_of_mentions / number_of_blocklists.
     @param returns: A dict of DomainBlocks keyed by domain
     """
-    merged = {}
+    merged = Blocklist('fediblockhole.merge_blocklists')
 
-    for key, blist in blocklists.items():
+    num_blocklists = len(blocklists)
-        log.debug(f"processing blocklist from: {key} ...")
+
-        for newblock in blist:
+    # Create a domain keyed list of blocks for each domain
-            domain = newblock.domain
+    domain_blocks = {}
-            # If the domain has two asterisks in it, it's obfuscated
+
-            # and we can't really use it, so skip it and do the next one
+    for bl in blocklists:
-            if '*' in domain:
+        for block in bl.values():
-                log.debug(f"Domain '{domain}' is obfuscated. Skipping it.")
+            if '*' in block.domain:
+                log.debug(f"Domain '{block.domain}' is obfuscated. Skipping it.")
                 continue
-
+            elif block.domain in domain_blocks:
-            elif domain in merged:
+                domain_blocks[block.domain].append(block)
-                log.debug(f"Overlapping block for domain {domain}. Merging...")
-                blockdata = apply_mergeplan(merged[domain], newblock, mergeplan)
-
             else:
-                # New block
+                domain_blocks[block.domain] = [block,]
-                blockdata = newblock
+
+    # Only merge items if `threshold` is met or exceeded
+    for domain in domain_blocks:
+        if threshold_type == 'count':
+            domain_threshold_level = len(domain_blocks[domain])
+        elif threshold_type == 'pct':
+            domain_threshold_level = len(domain_blocks[domain]) / num_blocklists * 100
+            # log.debug(f"domain threshold level: {domain_threshold_level}")
+        else:
+            raise ValueError(f"Unsupported threshold type '{threshold_type}'. Supported values are: 'count', 'pct'")
+
+        log.debug(f"Checking if {domain_threshold_level} >= {threshold} for {domain}")
+        if domain_threshold_level >= threshold:
+            # Add first block in the list to merged
+            block = domain_blocks[domain][0]
+            log.debug(f"Yes. Merging block: {block}")
+
+            # Merge the others with this record
+            for newblock in domain_blocks[domain][1:]:
+                block = apply_mergeplan(block, newblock, mergeplan)
+            merged.blocks[block.domain] = block
 
-            # end if
-            log.debug(f"blockdata is: {blockdata}")
-            merged[domain] = blockdata
-        # end for
     return merged
 
 def apply_mergeplan(oldblock: DomainBlock, newblock: DomainBlock, mergeplan: str='max') -> dict:
@@ -239,10 +264,10 @@ def apply_mergeplan(oldblock: DomainBlock, newblock: DomainBlock, mergeplan: str
     # How do we override an earlier block definition?
     if mergeplan in ['max', None]:
         # Use the highest block level found (the default)
-        log.debug(f"Using 'max' mergeplan.")
+        # log.debug(f"Using 'max' mergeplan.")
 
         if newblock.severity > oldblock.severity:
-            log.debug(f"New block severity is higher. Using that.")
+            # log.debug(f"New block severity is higher. Using that.")
             blockdata['severity'] = newblock.severity
         
         # For 'reject_media', 'reject_reports', and 'obfuscate' if
@@ -271,7 +296,7 @@ def apply_mergeplan(oldblock: DomainBlock, newblock: DomainBlock, mergeplan: str
     else:
         raise NotImplementedError(f"Mergeplan '{mergeplan}' not implemented.")
 
-    log.debug(f"Block severity set to {blockdata['severity']}")
+    # log.debug(f"Block severity set to {blockdata['severity']}")
 
     return DomainBlock(**blockdata)
 
@@ -357,17 +382,19 @@ def fetch_instance_blocklist(host: str, token: str=None, admin: bool=False,
 
     url = f"{scheme}://{host}{api_path}"
 
-    blocklist = []
+    blockdata = []
     link = True
-
     while link:
         response = requests.get(url, headers=headers, timeout=REQUEST_TIMEOUT)
         if response.status_code != 200:
             log.error(f"Cannot fetch remote blocklist: {response.content}")
             raise ValueError("Unable to fetch domain block list: %s", response)
 
-        blocklist.extend( parse_blocklist(response.content, parse_format, import_fields) )
+        # Each block of returned data is a JSON list of dicts
+        # so we parse them and append them to the fetched list
+        # of JSON data we need to parse.
 
+        blockdata.extend(json.loads(response.content.decode('utf-8')))
         # Parse the link header to find the next url to fetch
         # This is a weird and janky way of doing pagination but
         # hey nothing we can do about it we just have to deal
@@ -385,6 +412,8 @@ def fetch_instance_blocklist(host: str, token: str=None, admin: bool=False,
             urlstring, rel = next.split('; ')
             url = urlstring.strip('<').rstrip('>')
 
+    blocklist = parse_blocklist(blockdata, url, parse_format, import_fields)
+
     return blocklist
 
 def delete_block(token: str, host: str, id: int, scheme: str='https'):
@@ -474,13 +503,9 @@ def update_known_block(token: str, host: str, block: DomainBlock, scheme: str='h
     """Update an existing domain block with information in blockdict"""
     api_path = "/api/v1/admin/domain_blocks/"
 
-    try:
     id = block.id
     blockdata = block._asdict()
     del blockdata['id']
-    except KeyError:
-        import pdb
-        pdb.set_trace()
 
     url = f"{scheme}://{host}{api_path}{id}"
 
@@ -514,7 +539,7 @@ def add_block(token: str, host: str, blockdata: DomainBlock, scheme: str='https'
             
         raise ValueError(f"Something went wrong: {response.status_code}: {response.content}")
            
-def push_blocklist(token: str, host: str, blocklist: list[dict],
+def push_blocklist(token: str, host: str, blocklist: list[DomainBlock],
                     dryrun: bool=False,
                     import_fields: list=['domain', 'severity'],
                     max_followed_severity:BlockSeverity=BlockSeverity('silence'),
@@ -522,8 +547,7 @@ def push_blocklist(token: str, host: str, blocklist: list[dict],
                     ):
     """Push a blocklist to a remote instance.
     
-    Merging the blocklist with the existing list the instance has,
+    Updates existing entries if they exist, creates new blocks if they don't.
-    updating existing entries if they exist.
 
     @param token: The Bearer token for OAUTH API authentication
     @param host: The instance host, FQDN or IP
@@ -538,15 +562,16 @@ def push_blocklist(token: str, host: str, blocklist: list[dict],
     serverblocks = fetch_instance_blocklist(host, token, True, import_fields, scheme)
 
     # # Convert serverblocks to a dictionary keyed by domain name
-    knownblocks = {row.domain: row for row in serverblocks}
+    # knownblocks = {row.domain: row for row in serverblocks}
 
-    for newblock in blocklist:
+    for newblock in blocklist.values():
 
         log.debug(f"Processing block: {newblock}")
-        oldblock = knownblocks.get(newblock.domain, None)
+        if newblock.domain in serverblocks:
-        if oldblock:
             log.debug(f"Block already exists for {newblock.domain}, checking for differences...")
 
+            oldblock = serverblocks[newblock.domain]
+
             change_needed = is_change_needed(oldblock, newblock, import_fields)
 
             # Is the severity changing?
@@ -605,15 +630,14 @@ def load_config(configfile: str):
     conf = toml.load(configfile)
     return conf
 
-def save_intermediate_blocklist(
+def save_intermediate_blocklist(blocklist: Blocklist, filedir: str,
-    blocklist: list[dict], source: str,
-    filedir: str,
     export_fields: list=['domain','severity']):
     """Save a local copy of a blocklist we've downloaded
     """
     # Invent a filename based on the remote source
     # If the source was a URL, convert it to something less messy
     # If the source was a remote domain, just use the name of the domain
+    source = blocklist.origin
     log.debug(f"Saving intermediate blocklist from {source}")
     source = source.replace('/','-')
     filename = f"{source}.csv"
@@ -621,7 +645,7 @@ def save_intermediate_blocklist(
     save_blocklist_to_file(blocklist, filepath, export_fields)
 
 def save_blocklist_to_file(
-    blocklist: list[DomainBlock],
+    blocklist: Blocklist,
     filepath: str,
     export_fields: list=['domain','severity']):
     """Save a blocklist we've downloaded from a remote source
@@ -631,18 +655,22 @@ def save_blocklist_to_file(
     @param export_fields: Which fields to include in the export.
     """
     try:
-        blocklist = sorted(blocklist, key=lambda x: x.domain)
+        sorted_list = sorted(blocklist.blocks.items())
     except KeyError:
         log.error("Field 'domain' not found in blocklist.")
-        log.debug(f"blocklist is: {blocklist}")
+        log.debug(f"blocklist is: {sorted_list}")
+    except AttributeError:
+        log.error("Attribute error!")
+        import pdb
+        pdb.set_trace()
 
     log.debug(f"export fields: {export_fields}")
 
     with open(filepath, "w") as fp:
         writer = csv.DictWriter(fp, export_fields, extrasaction='ignore')
         writer.writeheader()
-        for item in blocklist:
+        for key, value in sorted_list:
-            writer.writerow(item._asdict())
+            writer.writerow(value)
 
 def augment_args(args, tomldata: str=None):
     """Augment commandline arguments with config file parameters
@@ -682,6 +710,12 @@ def augment_args(args, tomldata: str=None):
     if not args.mergeplan:
         args.mergeplan = conf.get('mergeplan', 'max')
 
+    if not args.merge_threshold:
+        args.merge_threshold = conf.get('merge_threshold', 0)
+
+    if not args.merge_threshold_type:
+        args.merge_threshold_type = conf.get('merge_threshold_type', 'count')
+
     args.blocklist_url_sources = conf.get('blocklist_url_sources', [])
     args.blocklist_instance_sources = conf.get('blocklist_instance_sources', [])
     args.allowlist_url_sources = conf.get('allowlist_url_sources', [])
@@ -703,6 +737,8 @@ def setup_argparse():
     ap.add_argument('-S', '--save-intermediate', dest="save_intermediate", action='store_true', help="Save intermediate blocklists we fetch to local files.")
     ap.add_argument('-D', '--savedir', dest="savedir", help="Directory path to save intermediate lists.")
     ap.add_argument('-m', '--mergeplan', choices=['min', 'max'], help="Set mergeplan.")
+    ap.add_argument('--merge-threshold', type=int, help="Merge threshold value")
+    ap.add_argument('--merge-threshold-type', choices=['count', 'pct'], help="Type of merge threshold to use.")
 
     ap.add_argument('-I', '--import-field', dest='import_fields', action='append', help="Extra blocklist fields to import.")
     ap.add_argument('-E', '--export-field', dest='export_fields', action='append', help="Extra blocklist fields to export.")

src/fediblockhole/blocklists.py

@@ -1,19 +1,47 @@
 """Parse various blocklist data formats
 """
-from typing import Iterable
-from .const import DomainBlock, BlockSeverity
-
 import csv
 import json
+from typing import Iterable
+from dataclasses import dataclass, field
+
+from .const import DomainBlock, BlockSeverity
 
 import logging
 log = logging.getLogger('fediblockhole')
 
+@dataclass
+class Blocklist:
+    """ A Blocklist object
+
+    A Blocklist is a list of DomainBlocks from an origin
+    """
+    origin: str = None
+    blocks: dict[str, DomainBlock] = field(default_factory=dict)
+
+    def __len__(self):
+        return len(self.blocks)
+
+    def __class_getitem__(cls, item):
+        return dict[str, DomainBlock]
+
+    def __getitem__(self, item):
+        return self.blocks[item]
+
+    def __iter__(self):
+        return self.blocks.__iter__()
+
+    def items(self):
+        return self.blocks.items()
+
+    def values(self):
+        return self.blocks.values()
+
 class BlocklistParser(object):
     """
     Base class for parsing blocklists
     """
-    preparse = False
+    do_preparse = False
 
     def __init__(self, import_fields: list=['domain', 'severity'], 
         max_severity: str='suspend'):
@@ -30,17 +58,18 @@ class BlocklistParser(object):
         """
         raise NotImplementedError
 
-    def parse_blocklist(self, blockdata) -> dict[DomainBlock]:
+    def parse_blocklist(self, blockdata, origin:str=None) -> Blocklist:
         """Parse an iterable of blocklist items
         @param blocklist: An Iterable of blocklist items
         @returns: A dict of DomainBlocks, keyed by domain
         """
-        if self.preparse:
+        if self.do_preparse:
             blockdata = self.preparse(blockdata)
 
-        parsed_list = []
+        parsed_list = Blocklist(origin)
         for blockitem in blockdata:
-            parsed_list.append(self.parse_item(blockitem))
+            block = self.parse_item(blockitem)
+            parsed_list.blocks[block.domain] = block
         return parsed_list
     
     def parse_item(self, blockitem) -> DomainBlock:
@@ -53,12 +82,13 @@ class BlocklistParser(object):
 
 class BlocklistParserJSON(BlocklistParser):
     """Parse a JSON formatted blocklist"""
-    preparse = True
+    do_preparse = True
 
     def preparse(self, blockdata) -> Iterable:
-        """Parse the blockdata as JSON
+        """Parse the blockdata as JSON if needed"""
-        """
+        if type(blockdata) == type(''):
             return json.loads(blockdata)
+        return blockdata
 
     def parse_item(self, blockitem: dict) -> DomainBlock:
         # Remove fields we don't want to import
@@ -102,7 +132,7 @@ class BlocklistParserCSV(BlocklistParser):
 
     The parser expects the CSV data to include a header with the field names.
     """
-    preparse = True
+    do_preparse = True
 
     def preparse(self, blockdata) -> Iterable:
         """Use a csv.DictReader to create an iterable from the blockdata
@@ -130,6 +160,24 @@ class BlocklistParserCSV(BlocklistParser):
             block.severity = self.max_severity
         return block
 
+class BlocklistParserMastodonCSV(BlocklistParserCSV):
+    """ Parse Mastodon CSV formatted blocklists
+
+    The Mastodon v4.1.x domain block CSV export prefixes its
+    field names with a '#' character because… reasons?
+    """
+    do_preparse = True
+
+    def parse_item(self, blockitem: dict) -> DomainBlock:
+        """Build a new blockitem dict with new un-#ed keys
+        """
+        newdict = {}
+        for key in blockitem:
+            newkey = key.lstrip('#')
+            newdict[newkey] = blockitem[key]
+
+        return super().parse_item(newdict)
+
 class RapidBlockParserCSV(BlocklistParserCSV):
     """ Parse RapidBlock CSV blocklists
 
@@ -193,6 +241,7 @@ def str2bool(boolstring: str) -> bool:
 
 FORMAT_PARSERS = {
     'csv': BlocklistParserCSV,
+    'mastodon_csv': BlocklistParserMastodonCSV,
     'json': BlocklistParserJSON,
     'mastodon_api_public': BlocklistParserMastodonAPIPublic,
     'rapidblock.csv': RapidBlockParserCSV,
@@ -202,11 +251,13 @@ FORMAT_PARSERS = {
 # helper function to select the appropriate Parser
 def parse_blocklist(
     blockdata,
+    origin,
     format="csv",
     import_fields: list=['domain', 'severity'],
     max_severity: str='suspend'):
     """Parse a blocklist in the given format
     """
-    parser = FORMAT_PARSERS[format](import_fields, max_severity)
     log.debug(f"parsing {format} blocklist with import_fields: {import_fields}...")
-    return parser.parse_blocklist(blockdata)
+
+    parser = FORMAT_PARSERS[format](import_fields, max_severity)
+    return parser.parse_blocklist(blockdata, origin)

tests/test_allowlist.py

@@ -4,6 +4,7 @@ import pytest
 
 from util import shim_argparse
 from fediblockhole.const import DomainBlock
+from fediblockhole.blocklists import Blocklist
 from fediblockhole import fetch_allowlists, apply_allowlists
 
 def test_cmdline_allow_removes_domain():
@@ -11,16 +12,12 @@ def test_cmdline_allow_removes_domain():
     """
     conf = shim_argparse(['-A', 'removeme.org'])
 
-    merged = {
+    merged = Blocklist('test_allowlist.merged', {
         'example.org': DomainBlock('example.org'),
         'example2.org': DomainBlock('example2.org'),
         'removeme.org': DomainBlock('removeme.org'),
         'keepblockingme.org': DomainBlock('keepblockingme.org'),
-    }
+    })
-
-    # allowlists = {
-    #     'testlist': [ DomainBlock('removeme.org', 'noop'), ]
-    # }
 
     merged = apply_allowlists(merged, conf, {})
 
@@ -32,16 +29,18 @@ def test_allowlist_removes_domain():
     """
     conf = shim_argparse()
 
-    merged = {
+    merged = Blocklist('test_allowlist.merged', {
         'example.org': DomainBlock('example.org'),
         'example2.org': DomainBlock('example2.org'),
         'removeme.org': DomainBlock('removeme.org'),
         'keepblockingme.org': DomainBlock('keepblockingme.org'),
-    }
+    })
 
-    allowlists = {
+    allowlists = [
-        'testlist': [ DomainBlock('removeme.org', 'noop'), ]
+        Blocklist('test_allowlist', {
-    }
+            'removeme.org': DomainBlock('removeme.org', 'noop'),
+            })
+    ]
     
     merged = apply_allowlists(merged, conf, allowlists)
 
@@ -53,19 +52,19 @@ def test_allowlist_removes_tld():
     """
     conf = shim_argparse()
 
-    merged = {
+    merged = Blocklist('test_allowlist.merged', {
         '.cf': DomainBlock('.cf'),
         'example.org': DomainBlock('example.org'),
         '.tk': DomainBlock('.tk'),
         'keepblockingme.org': DomainBlock('keepblockingme.org'),
-    }
+    })
 
-    allowlists = {
+    allowlists = [
-        'list1': [
+        Blocklist('test_allowlist.list1', {
-            DomainBlock('.cf', 'noop'), 
+        '.cf': DomainBlock('.cf', 'noop'),
-            DomainBlock('.tk', 'noop'), 
+        '.tk': DomainBlock('.tk', 'noop'), 
+        })
     ]
-    }
     
     merged = apply_allowlists(merged, conf, allowlists)
 

tests/test_configfile.py

@@ -49,3 +49,33 @@ allowlist_url_sources = [ { url='file:///path/to/allowlist', format='csv'} ]
         'url': 'file:///path/to/allowlist',
         'format': 'csv',
         }]
+
+def test_set_merge_thresold_default():
+    tomldata = """
+"""
+    args = shim_argparse([], tomldata)
+
+    assert args.mergeplan == 'max'
+    assert args.merge_threshold_type == 'count'
+
+def test_set_merge_thresold_count():
+    tomldata = """# Add a merge threshold
+merge_threshold_type = 'count'
+merge_threshold = 2
+"""
+    args = shim_argparse([], tomldata)
+
+    assert args.mergeplan == 'max'
+    assert args.merge_threshold_type == 'count'
+    assert args.merge_threshold == 2
+
+def test_set_merge_thresold_pct():
+    tomldata = """# Add a merge threshold
+merge_threshold_type = 'pct'
+merge_threshold = 35
+"""
+    args = shim_argparse([], tomldata)
+
+    assert args.mergeplan == 'max'
+    assert args.merge_threshold_type == 'pct'
+    assert args.merge_threshold == 35

tests/test_merge_thresholds.py

@@ -0,0 +1,153 @@
+"""Test merge with thresholds
+"""
+
+from fediblockhole.blocklists import Blocklist, parse_blocklist
+from fediblockhole import merge_blocklists, apply_mergeplan
+
+from fediblockhole.const import SeverityLevel, DomainBlock
+
+datafile01 = "data-suspends-01.csv"
+datafile02 = "data-silences-01.csv"
+datafile03 = "data-noop-01.csv"
+
+import_fields = [
+    'domain',
+    'severity',
+    'public_comment',
+    'private_comment',
+    'reject_media',
+    'reject_reports',
+    'obfuscate'
+]
+
+def load_test_blocklist_data(datafiles):
+
+    blocklists = []
+
+    for df in datafiles:
+        with open(df) as fp:
+            data = fp.read()
+            bl = parse_blocklist(data, df, 'csv', import_fields)
+            blocklists.append(bl)
+    
+    return blocklists
+
+def test_mergeplan_count_2():
+    """Only merge a block if present in 2 or more lists
+    """
+
+    bl_1 = Blocklist('test01', {
+        'onemention.example.org': DomainBlock('onemention.example.org', 'suspend', '', '', True, True, True),
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        })
+
+    bl_2 = Blocklist('test2', {
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_3 = Blocklist('test3', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    ml = merge_blocklists([bl_1, bl_2, bl_3], 'max', threshold=2)
+
+    assert 'onemention.example.org' not in ml
+    assert 'twomention.example.org' in ml
+    assert 'threemention.example.org' in ml
+
+def test_mergeplan_count_3():
+    """Only merge a block if present in 3 or more lists
+    """
+
+    bl_1 = Blocklist('test01', {
+        'onemention.example.org': DomainBlock('onemention.example.org', 'suspend', '', '', True, True, True),
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        })
+
+    bl_2 = Blocklist('test2', {
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_3 = Blocklist('test3', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    ml = merge_blocklists([bl_1, bl_2, bl_3], 'max', threshold=3)
+
+    assert 'onemention.example.org' not in ml
+    assert 'twomention.example.org' not in ml
+    assert 'threemention.example.org' in ml
+
+def test_mergeplan_pct_30():
+    """Only merge a block if present in 2 or more lists
+    """
+
+    bl_1 = Blocklist('test01', {
+        'onemention.example.org': DomainBlock('onemention.example.org', 'suspend', '', '', True, True, True),
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+
+        })
+
+    bl_2 = Blocklist('test2', {
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_3 = Blocklist('test3', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_4 = Blocklist('test4', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    ml = merge_blocklists([bl_1, bl_2, bl_3, bl_4], 'max', threshold=30, threshold_type='pct')
+
+    assert 'onemention.example.org' not in ml
+    assert 'twomention.example.org' in ml
+    assert 'threemention.example.org' in ml
+    assert 'fourmention.example.org' in ml
+
+def test_mergeplan_pct_55():
+    """Only merge a block if present in 2 or more lists
+    """
+
+    bl_1 = Blocklist('test01', {
+        'onemention.example.org': DomainBlock('onemention.example.org', 'suspend', '', '', True, True, True),
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+
+        })
+
+    bl_2 = Blocklist('test2', {
+        'twomention.example.org': DomainBlock('twomention.example.org', 'suspend', '', '', True, True, True),
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_3 = Blocklist('test3', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    bl_4 = Blocklist('test4', {
+        'threemention.example.org': DomainBlock('threemention.example.org', 'suspend', '', '', True, True, True),
+        'fourmention.example.org': DomainBlock('fourmention.example.org', 'suspend', '', '', True, True, True),
+    })
+
+    ml = merge_blocklists([bl_1, bl_2, bl_3, bl_4], 'max', threshold=55, threshold_type='pct')
+
+    assert 'onemention.example.org' not in ml
+    assert 'twomention.example.org' not in ml
+    assert 'threemention.example.org' in ml
+    assert 'fourmention.example.org' in ml

tests/test_mergeplan.py

@@ -1,7 +1,7 @@
 """Various mergeplan tests
 """
 
-from fediblockhole.blocklist_parser import parse_blocklist
+from fediblockhole.blocklists import parse_blocklist
 from fediblockhole import merge_blocklists, merge_comments, apply_mergeplan
 
 from fediblockhole.const import SeverityLevel, DomainBlock
@@ -22,20 +22,19 @@ import_fields = [
 
 def load_test_blocklist_data(datafiles):
 
-    blocklists = {}
+    blocklists = []
 
     for df in datafiles:
         with open(df) as fp:
             data = fp.read()
-            bl = parse_blocklist(data, 'csv', import_fields)
+            bl = parse_blocklist(data, df, 'csv', import_fields)
-            blocklists[df] = bl
+            blocklists.append(bl)
     
     return blocklists
 
 def test_mergeplan_max():
     """Test 'max' mergeplan"""
     blocklists = load_test_blocklist_data([datafile01, datafile02])
-
     bl = merge_blocklists(blocklists, 'max')
     assert len(bl) == 13
 

tests/test_parser_csv.py

@@ -1,22 +1,24 @@
 """Tests of the CSV parsing
 """
 
-from fediblockhole.blocklist_parser import BlocklistParserCSV, parse_blocklist
+from fediblockhole.blocklists import BlocklistParserCSV, parse_blocklist
-from fediblockhole.const import DomainBlock, BlockSeverity, SeverityLevel
+from fediblockhole.const import SeverityLevel
 
 
 def test_single_line():
     csvdata = "example.org"
+    origin = "csvfile"
 
     parser = BlocklistParserCSV()
-    bl = parser.parse_blocklist(csvdata)
+    bl = parser.parse_blocklist(csvdata, origin)
     assert len(bl) == 0
 
 def test_header_only():
     csvdata = "domain,severity,public_comment"
+    origin = "csvfile"
 
     parser = BlocklistParserCSV()
-    bl = parser.parse_blocklist(csvdata)
+    bl = parser.parse_blocklist(csvdata, origin)
     assert len(bl) == 0
 
 def test_2_blocks():
@@ -24,12 +26,13 @@ def test_2_blocks():
 example.org,silence
 example2.org,suspend
 """
+    origin = "csvfile"
 
     parser = BlocklistParserCSV()
-    bl = parser.parse_blocklist(csvdata)
+    bl = parser.parse_blocklist(csvdata, origin)
 
     assert len(bl) == 2
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
 
 def test_4_blocks():
     csvdata = """domain,severity,public_comment
@@ -38,20 +41,21 @@ example2.org,suspend,"test 2"
 example3.org,noop,"test 3"
 example4.org,suspend,"test 4"
 """
+    origin = "csvfile"
 
     parser = BlocklistParserCSV()
-    bl = parser.parse_blocklist(csvdata)
+    bl = parser.parse_blocklist(csvdata, origin)
 
     assert len(bl) == 4
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
-    assert bl[1].domain == 'example2.org'
+    assert 'example2.org' in bl
-    assert bl[2].domain == 'example3.org'
+    assert 'example3.org' in bl
-    assert bl[3].domain == 'example4.org'
+    assert 'example4.org' in bl
 
-    assert bl[0].severity.level == SeverityLevel.SILENCE
+    assert bl['example.org'].severity.level == SeverityLevel.SILENCE
-    assert bl[1].severity.level == SeverityLevel.SUSPEND
+    assert bl['example2.org'].severity.level == SeverityLevel.SUSPEND
-    assert bl[2].severity.level == SeverityLevel.NONE
+    assert bl['example3.org'].severity.level == SeverityLevel.NONE
-    assert bl[3].severity.level == SeverityLevel.SUSPEND
+    assert bl['example4.org'].severity.level == SeverityLevel.SUSPEND
 
 def test_ignore_comments():
     csvdata = """domain,severity,public_comment,private_comment
@@ -60,18 +64,18 @@ example2.org,suspend,"test 2","ignote me also"
 example3.org,noop,"test 3","and me"
 example4.org,suspend,"test 4","also me"
 """
+    origin = "csvfile"
 
     parser = BlocklistParserCSV()
-    bl = parser.parse_blocklist(csvdata)
+    bl = parser.parse_blocklist(csvdata, origin)
 
     assert len(bl) == 4
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
-    assert bl[1].domain == 'example2.org'
+    assert 'example2.org' in bl
-    assert bl[2].domain == 'example3.org'
+    assert 'example3.org' in bl
-    assert bl[3].domain == 'example4.org'
+    assert 'example4.org' in bl
 
-    assert bl[0].public_comment == ''
+    assert bl['example.org'].public_comment == ''
-    assert bl[0].private_comment == ''
+    assert bl['example.org'].private_comment == ''
-
+    assert bl['example3.org'].public_comment == ''
-    assert bl[2].public_comment == ''
+    assert bl['example4.org'].private_comment == ''
-    assert bl[2].private_comment == ''

tests/test_parser_csv_mastodon.py

@@ -0,0 +1,81 @@
+"""Tests of the CSV parsing
+"""
+
+from fediblockhole.blocklists import BlocklistParserMastodonCSV
+from fediblockhole.const import SeverityLevel
+
+
+def test_single_line():
+    csvdata = "example.org"
+    origin = "csvfile"
+
+    parser = BlocklistParserMastodonCSV()
+    bl = parser.parse_blocklist(csvdata, origin)
+    assert len(bl) == 0
+
+def test_header_only():
+    csvdata = "#domain,#severity,#public_comment"
+    origin = "csvfile"
+
+    parser = BlocklistParserMastodonCSV()
+    bl = parser.parse_blocklist(csvdata, origin)
+    assert len(bl) == 0
+
+def test_2_blocks():
+    csvdata = """domain,severity
+example.org,silence
+example2.org,suspend
+"""
+    origin = "csvfile"
+
+    parser = BlocklistParserMastodonCSV()
+    bl = parser.parse_blocklist(csvdata, origin)
+
+    assert len(bl) == 2
+    assert 'example.org' in bl
+
+def test_4_blocks():
+    csvdata = """domain,severity,public_comment
+example.org,silence,"test 1"
+example2.org,suspend,"test 2"
+example3.org,noop,"test 3"
+example4.org,suspend,"test 4"
+"""
+    origin = "csvfile"
+
+    parser = BlocklistParserMastodonCSV()
+    bl = parser.parse_blocklist(csvdata, origin)
+
+    assert len(bl) == 4
+    assert 'example.org' in bl
+    assert 'example2.org' in bl
+    assert 'example3.org' in bl
+    assert 'example4.org' in bl
+
+    assert bl['example.org'].severity.level == SeverityLevel.SILENCE
+    assert bl['example2.org'].severity.level == SeverityLevel.SUSPEND
+    assert bl['example3.org'].severity.level == SeverityLevel.NONE
+    assert bl['example4.org'].severity.level == SeverityLevel.SUSPEND
+
+def test_ignore_comments():
+    csvdata = """domain,severity,public_comment,private_comment
+example.org,silence,"test 1","ignore me"
+example2.org,suspend,"test 2","ignote me also"
+example3.org,noop,"test 3","and me"
+example4.org,suspend,"test 4","also me"
+"""
+    origin = "csvfile"
+
+    parser = BlocklistParserMastodonCSV()
+    bl = parser.parse_blocklist(csvdata, origin)
+
+    assert len(bl) == 4
+    assert 'example.org' in bl
+    assert 'example2.org' in bl
+    assert 'example3.org' in bl
+    assert 'example4.org' in bl
+
+    assert bl['example.org'].public_comment == ''
+    assert bl['example.org'].private_comment == ''
+    assert bl['example3.org'].public_comment == ''
+    assert bl['example4.org'].private_comment == ''

tests/test_parser_json.py

@@ -1,8 +1,8 @@
 """Tests of the CSV parsing
 """
 
-from fediblockhole.blocklist_parser import BlocklistParserJSON, parse_blocklist
+from fediblockhole.blocklists import BlocklistParserJSON, parse_blocklist
-from fediblockhole.const import DomainBlock, BlockSeverity, SeverityLevel
+from fediblockhole.const import SeverityLevel
 
 datafile = 'data-mastodon.json'
 
@@ -14,33 +14,32 @@ def test_json_parser():
 
     data = load_data()
     parser = BlocklistParserJSON()
-    bl = parser.parse_blocklist(data)
+    bl = parser.parse_blocklist(data, 'test_json')
 
     assert len(bl) == 10
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
-    assert bl[1].domain == 'example2.org'
+    assert 'example2.org' in bl
-    assert bl[2].domain == 'example3.org'
+    assert 'example3.org' in bl
-    assert bl[3].domain == 'example4.org'
+    assert 'example4.org' in bl
 
-    assert bl[0].severity.level == SeverityLevel.SUSPEND
+    assert bl['example.org'].severity.level == SeverityLevel.SUSPEND
-    assert bl[1].severity.level == SeverityLevel.SILENCE
+    assert bl['example2.org'].severity.level == SeverityLevel.SILENCE
-    assert bl[2].severity.level == SeverityLevel.SUSPEND
+    assert bl['example3.org'].severity.level == SeverityLevel.SUSPEND
-    assert bl[3].severity.level == SeverityLevel.NONE
+    assert bl['example4.org'].severity.level == SeverityLevel.NONE
 
 def test_ignore_comments():
 
     data = load_data()
     parser = BlocklistParserJSON()
-    bl = parser.parse_blocklist(data)
+    bl = parser.parse_blocklist(data, 'test_json')
 
     assert len(bl) == 10
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
-    assert bl[1].domain == 'example2.org'
+    assert 'example2.org' in bl
-    assert bl[2].domain == 'example3.org'
+    assert 'example3.org' in bl
-    assert bl[3].domain == 'example4.org'
+    assert 'example4.org' in bl
 
-    assert bl[0].public_comment == ''
+    assert bl['example.org'].public_comment == ''
-    assert bl[0].private_comment == ''
+    assert bl['example.org'].private_comment == ''
-
+    assert bl['example3.org'].public_comment == ''
-    assert bl[2].public_comment == ''
+    assert bl['example4.org'].private_comment == ''
-    assert bl[2].private_comment == ''

tests/test_parser_rapidblockcsv.py

@@ -1,7 +1,7 @@
 """Tests of the Rapidblock CSV parsing
 """
 
-from fediblockhole.blocklist_parser import RapidBlockParserCSV, parse_blocklist
+from fediblockhole.blocklists import RapidBlockParserCSV, parse_blocklist
 from fediblockhole.const import DomainBlock, BlockSeverity, SeverityLevel
 
 csvdata = """example.org\r\nsubdomain.example.org\r\nanotherdomain.org\r\ndomain4.org\r\n"""
@@ -11,13 +11,13 @@ def test_basic_rapidblock():
 
     bl = parser.parse_blocklist(csvdata)
     assert len(bl) == 4
-    assert bl[0].domain == 'example.org'
+    assert 'example.org' in bl
-    assert bl[1].domain == 'subdomain.example.org'
+    assert 'subdomain.example.org' in bl
-    assert bl[2].domain == 'anotherdomain.org'
+    assert 'anotherdomain.org' in bl
-    assert bl[3].domain == 'domain4.org'
+    assert 'domain4.org' in bl
 
 def test_severity_is_suspend():
     bl = parser.parse_blocklist(csvdata)
 
-    for block in bl:
+    for block in bl.values():
         assert block.severity.level == SeverityLevel.SUSPEND

tests/test_parser_rapidblockjson.py

@@ -1,6 +1,6 @@
 """Test parsing the RapidBlock JSON format
 """
-from fediblockhole.blocklist_parser import parse_blocklist
+from fediblockhole.blocklists import parse_blocklist
 
 from fediblockhole.const import SeverityLevel
 
@@ -9,26 +9,26 @@ rapidblockjson = "data-rapidblock.json"
 def test_parse_rapidblock_json():
     with open(rapidblockjson) as fp:
         data = fp.read()
-        bl = parse_blocklist(data, 'rapidblock.json')
+        bl = parse_blocklist(data, 'pytest', 'rapidblock.json')
 
-        assert bl[0].domain == '101010.pl'
+        assert '101010.pl' in bl
-        assert bl[0].severity.level == SeverityLevel.SUSPEND
+        assert bl['101010.pl'].severity.level == SeverityLevel.SUSPEND
-        assert bl[0].public_comment == ''
+        assert bl['101010.pl'].public_comment == ''
 
-        assert bl[10].domain == 'berserker.town'
+        assert 'berserker.town' in bl
-        assert bl[10].severity.level == SeverityLevel.SUSPEND
+        assert bl['berserker.town'].severity.level == SeverityLevel.SUSPEND
-        assert bl[10].public_comment == ''
+        assert bl['berserker.town'].public_comment == ''
-        assert bl[10].private_comment == ''
+        assert bl['berserker.town'].private_comment == ''
 
 def test_parse_with_comments():
     with open(rapidblockjson) as fp:
         data = fp.read()
-        bl = parse_blocklist(data, 'rapidblock.json', ['domain', 'severity', 'public_comment', 'private_comment'])
+        bl = parse_blocklist(data, 'pytest', 'rapidblock.json', ['domain', 'severity', 'public_comment', 'private_comment'])
 
-        assert bl[0].domain == '101010.pl'
+        assert '101010.pl' in bl
-        assert bl[0].severity.level == SeverityLevel.SUSPEND
+        assert bl['101010.pl'].severity.level == SeverityLevel.SUSPEND
-        assert bl[0].public_comment == 'cryptomining javascript, white supremacy'
+        assert bl['101010.pl'].public_comment == 'cryptomining javascript, white supremacy'
 
-        assert bl[10].domain == 'berserker.town'
+        assert 'berserker.town' in bl
-        assert bl[10].severity.level == SeverityLevel.SUSPEND
+        assert bl['berserker.town'].severity.level == SeverityLevel.SUSPEND
-        assert bl[10].public_comment == 'freeze peach'
+        assert bl['berserker.town'].public_comment == 'freeze peach'