epstein-files 1.0.2__py3-none-any.whl → 1.0.4__py3-none-any.whl

epstein_files/__init__.py

@@ -4,7 +4,7 @@ Reformat Epstein text message files for readability and count email senders.
 For use with iMessage log files from https://drive.google.com/drive/folders/1hTNH5woIRio578onLGElkTWofUSWRoH_
 
 Install: 'poetry install'
-    Run: 'EPSTEIN_DOCS_DIR=/path/to/TXT ./generate.py'
+    Run: 'EPSTEIN_DOCS_DIR=/path/to/TXT epstein_generate'
 """
 from sys import exit
 
@@ -87,6 +87,9 @@ def epstein_search():
             console.line()
 
             if args.whole_file:
+                if isinstance(search_result.document, Email):
+                    search_result.document.truncation_allowed = False
+
                 console.print(search_result.document)
             else:
                 console.print(search_result.document.description_panel())

epstein_files/documents/document.py

@@ -15,7 +15,7 @@ from epstein_files.util.constant.names import *
 from epstein_files.util.constant.strings import *
 from epstein_files.util.constant.urls import *
 from epstein_files.util.constants import ALL_FILE_CONFIGS, FALLBACK_TIMESTAMP
-from epstein_files.util.data import collapse_newlines, date_str, iso_timestamp, listify, patternize, without_nones
+from epstein_files.util.data import collapse_newlines, date_str, iso_timestamp, listify, patternize, without_falsey
 from epstein_files.util.doc_cfg import EmailCfg, DocCfg, Metadata, TextCfg
 from epstein_files.util.env import args
 from epstein_files.util.file_helper import (DOCS_DIR, file_stem_for_id, extract_file_id, file_size,
@@ -159,7 +159,7 @@ class Document:
         if hint_msg:
             hints.append(highlighter(Text(hint_msg, style='white dim italic')))
 
-        return without_nones(hints)
+        return without_falsey(hints)
 
     def info_txt(self) -> Text | None:
         """Secondary info about this file (recipients, level of certainty, etc). Overload in subclasses."""
@@ -356,6 +356,11 @@ class Document:
         for f in tmpfiles:
             f.unlink()
 
+    @staticmethod
+    def known_author_count(docs: Sequence['Document']) -> int:
+        """Count of how many Document objects have an author attribution."""
+        return len([doc for doc in docs if doc.author])
+
     @staticmethod
     def sort_by_timestamp(docs: Sequence['DocumentType']) -> list['DocumentType']:
         return sorted(docs, key=lambda doc: doc.sort_key())

epstein_files/documents/email.py

@@ -30,7 +30,6 @@ BAD_LINE_REGEX = re.compile(r'^(>;?|\d{1,2}|Classification: External Communicati
 DETECT_EMAIL_REGEX = re.compile(r'^(.*\n){0,2}From:')
 LINK_LINE_REGEX = re.compile(f"^(> )?htt")
 QUOTED_REPLY_LINE_REGEX = re.compile(r'wrote:\n', re.IGNORECASE)
-REPLY_SPLITTERS = [f"{field}:" for field in FIELD_NAMES] + ['********************************']
 REPLY_TEXT_REGEX = re.compile(rf"^(.*?){REPLY_LINE_PATTERN}", re.DOTALL | re.IGNORECASE | re.MULTILINE)
 
 BAD_TIMEZONE_REGEX = re.compile(fr'\((UTC|GMT\+\d\d:\d\d)\)|{REDACTED}')
@@ -39,10 +38,16 @@ TIMESTAMP_LINE_REGEX = re.compile(r"\d+:\d+")
 
 SUPPRESS_LOGS_FOR_AUTHORS = ['Undisclosed recipients:', 'undisclosed-recipients:', 'Multiple Senders Multiple Senders']
 REWRITTEN_HEADER_MSG = "(janky OCR header fields were prettified, check source if something seems off)"
+IS_JUNK_MAIL = 'is_junk_mail'
 MAX_CHARS_TO_PRINT = 4000
 MAX_NUM_HEADER_LINES = 14
 MAX_QUOTED_REPLIES = 2
 
+REPLY_SPLITTERS = [f"{field}:" for field in FIELD_NAMES] + [
+    '********************************',
+    'Begin forwarded message',
+]
+
 OCR_REPAIRS: dict[str | re.Pattern, str] = {
     re.compile(r'grnail\.com'): 'gmail.com',
     re.compile(r"^(From|To)(: )?[_1.]{5,}", re.MULTILINE): rf"\1: {REDACTED}",  # Redacted email addresses
@@ -119,6 +124,7 @@ EMAIL_SIGNATURE_REGEXES = {
 # Invalid for links to EpsteinWeb
 JUNK_EMAILERS = [
     'asmallworld@travel.asmallworld.net',
+    "digest-noreply@quora.com",
     'editorialstaff@flipboard.com',
     'How To Academy',
     'Jokeland',
@@ -126,9 +132,13 @@ JUNK_EMAILERS = [
     'Saved by Internet Explorer 11',
 ]
 
-TRUNCATE_ALL_EMAILS_FROM = JUNK_EMAILERS + [
-    'Alan S Halperin',
+MAILING_LISTS = [
+    INTELLIGENCE_SQUARED,
     'middle.east.update@hotmail.com',
+]
+
+TRUNCATE_ALL_EMAILS_FROM = JUNK_EMAILERS + MAILING_LISTS + [
+    'Alan S Halperin',
     'Mitchell Bard',
     'Skip Rimer',
 ]
@@ -281,7 +291,7 @@ SELF_EMAILS_FILE_IDS = [
 ]
 
 METADATA_FIELDS = [
-    'is_junk_mail',
+    IS_JUNK_MAIL,
     'recipients',
     'sent_from_device',
 ]
@@ -294,7 +304,6 @@ class Email(Communication):
         actual_text (str) - best effort at the text actually sent in this email, excluding quoted replies and forwards
         config (EmailCfg | None) - manual config for this email (if it exists)
         header (EmailHeader) - header data extracted from the text (from/to/sent/subject etc)
-        is_junk_mail (bool) - True if this is junk mail
         recipients (list[str | None]) - who this email was sent to
         sent_from_device (str | None) - "Sent from my iPhone" style signature (if it exists)
         signature_substitution_counts (dict[str, int]) - count of how many times a signature was replaced with <...snipped...> for each participant
@@ -302,17 +311,16 @@ class Email(Communication):
     actual_text: str = field(init=False)
     config: EmailCfg | None = None
     header: EmailHeader = field(init=False)
-    is_junk_mail: bool = False
     recipients: list[str | None] = field(default_factory=list)
     sent_from_device: str | None = None
     signature_substitution_counts: dict[str, int] = field(default_factory=dict)  # defaultdict breaks asdict :(
+    truncation_allowed: bool = True
 
     # For logging how many headers we prettified while printing, kind of janky
     rewritten_header_ids: ClassVar[set[str]] = set([])
 
     def __post_init__(self):
         super().__post_init__()
-        self.is_junk_mail = self.author in JUNK_EMAILERS
 
         if self.config and self.config.recipients:
             self.recipients = cast(list[str | None], self.config.recipients)
@@ -331,9 +339,17 @@ class Email(Communication):
         txt = Text("OCR text of email from ", style='grey46').append(self.author_txt).append(' to ')
         return txt.append(self._recipients_txt()).append(highlighter(f" probably sent at {self.timestamp}"))
 
+    def is_fwded_article(self) -> bool:
+        return bool(self.config and self.config.is_fwded_article)
+
+    def is_junk_mail(self) -> bool:
+        return self.author in JUNK_EMAILERS or self.author in MAILING_LISTS
+
     def metadata(self) -> Metadata:
+        local_metadata = asdict(self)
+        local_metadata[IS_JUNK_MAIL] = self.is_junk_mail()
         metadata = super().metadata()
-        metadata.update({k: v for k, v in asdict(self).items() if v and k in METADATA_FIELDS})
+        metadata.update({k: v for k, v in local_metadata.items() if v and k in METADATA_FIELDS})
         return metadata
 
     def subject(self) -> str:
@@ -352,17 +368,18 @@ class Email(Communication):
         """The text that comes before likely quoted replies and forwards etc."""
         if self.config and self.config.actual_text is not None:
             return self.config.actual_text
+
+        text = '\n'.join(self.text.split('\n')[self.header.num_header_rows:]).strip()
+
+        if self.config and self.config.fwded_text_after:
+            return text.split(self.config.fwded_text_after)[0].strip()
         elif self.header.num_header_rows == 0:
             return self.text
 
-        text = '\n'.join(self.text.split('\n')[self.header.num_header_rows:]).strip()
         reply_text_match = REPLY_TEXT_REGEX.search(text)
         # logger.info(f"Raw text:\n" + self.top_lines(20) + '\n\n')
         # logger.info(f"With header removed:\n" + text[0:500] + '\n\n')
 
-        if self.file_id in ['024624']:  # This email starts with "On September 14th"
-            return text.split('On Tue, May 14')[0].strip()
-
         if reply_text_match:
             actual_num_chars = len(reply_text_match.group(1))
             actual_text_pct = f"{(100 * float(actual_num_chars) / len(text)):.1f}%"
@@ -555,6 +572,9 @@ class Email(Communication):
             self._merge_lines(3, 5)
         elif self.file_id == '028931':
             self._merge_lines(3, 6)
+        elif self.file_id == '013415':
+            for _i in range(2):
+                self._merge_lines(4)
         elif self.file_id in ['033568']:
             for _i in range(5):
                 self._merge_lines(5)
@@ -637,7 +657,7 @@ class Email(Communication):
             num_chars = quote_cutoff
 
         # Truncate long emails but leave a note explaining what happened w/link to source document
-        if len(text) > num_chars:
+        if len(text) > num_chars and self.truncation_allowed:
             text = text[0:num_chars]
             doc_link_markup = epstein_media_doc_link_markup(self.url_slug, self.author_style)
             trim_note = f"<...trimmed to {num_chars} characters of {self.length}, read the rest at {doc_link_markup}...>"

epstein_files/documents/imessage/text_message.py

@@ -4,7 +4,7 @@ from datetime import datetime
 
 from rich.text import Text
 
-from epstein_files.util.constant.names import JEFFREY_EPSTEIN, ANTHONY_SCARAMUCCI, STEVE_BANNON, UNKNOWN
+from epstein_files.util.constant.names import JEFFREY_EPSTEIN, ANTHONY_SCARAMUCCI, CELINA_DUBIN, EVA, STEVE_BANNON, UNKNOWN
 from epstein_files.util.data import extract_last_name
 from epstein_files.util.highlighted_group import get_style_for_name
 from epstein_files.util.logging import logger
@@ -19,17 +19,18 @@ DISPLAY_LAST_NAME_ONLY = [
     STEVE_BANNON,
 ]
 
-UNKNOWN_TEXTERS = [
-    '+16463880059',
-    '+13108737937',
-    '+13108802851',
-]
+PHONE_NUMBER_MAPPING = {
+    '+19174393646': ANTHONY_SCARAMUCCI,
+    '+13109906526': STEVE_BANNON,
+    '+16463880059': EVA,
+    '+13108737937': CELINA_DUBIN,
+    '+13108802851': STEVE_BANNON,
+
+}
 
 TEXTER_MAPPING = {
     'e:': JEFFREY_EPSTEIN,
     'e:jeeitunes@gmail.com': JEFFREY_EPSTEIN,
-    '+19174393646': ANTHONY_SCARAMUCCI,
-    '+13109906526': STEVE_BANNON,
 }
 
 
@@ -37,7 +38,7 @@ TEXTER_MAPPING = {
 class TextMessage:
     """Class representing a single iMessage text message."""
     author: str | None
-    author_str: str = field(init=False)
+    author_str: str | None = None
     id_confirmed: bool = False
     text: str
     timestamp_str: str
@@ -47,14 +48,10 @@ class TextMessage:
 
         if self.author is None:
             self.author_str = UNKNOWN
-        elif self.author in UNKNOWN_TEXTERS:
-            logger.warning(f"Bad text from '{self.author}': \"{self.text}\"")
-            self.author_str = self.author
-            self.author = None  # TODO: this shouldn't be happening; we still know the author...
         elif self.author in DISPLAY_LAST_NAME_ONLY:
             self.author_str = extract_last_name(self.author)
         else:
-            self.author_str = self.author
+            self.author_str = self.author_str or self.author
 
         if not self.id_confirmed and self.author is not None and self.author != JEFFREY_EPSTEIN:
             self.author_str = self.author + ' (?)'
@@ -87,7 +84,6 @@ class TextMessage:
         return msg_txt
 
     def __rich__(self) -> Text:
-        # TODO: Workaround for phone numbers that sucks
         author_style = get_style_for_name(self.author_str if self.author_str.startswith('+') else self.author)
         author_txt = Text(self.author_str, style=author_style)
         timestamp_txt = Text(f"[{self.timestamp_str}]", style=TIMESTAMP_STYLE).append(' ')

epstein_files/documents/messenger_log.py

@@ -44,17 +44,8 @@ class MessengerLog(Communication):
 
     def messages(self) -> list[TextMessage]:
         """Lazily evaluated accessor for self._messages."""
-        if len(self._messages) == 0:
-            self._messages = [
-                TextMessage(
-                    # If the Sender: is redacted that means it's from self.author
-                    author=REDACTED_AUTHOR_REGEX.sub('', match.group(1).strip()) or self.author,
-                    id_confirmed=not self.is_attribution_uncertain(),
-                    text=match.group(4).strip(),
-                    timestamp_str=match.group(2).strip(),
-                )
-                for match in MSG_REGEX.finditer(self.text)
-            ]
+        if not self._messages:
+            self._messages = [self._build_message(match) for match in MSG_REGEX.finditer(self.text)]
 
         return self._messages
 
@@ -70,6 +61,19 @@ class MessengerLog(Communication):
     def _border_style(self) -> str:
         return self.author_style
 
+    def _build_message(self, match: re.Match) -> TextMessage:
+        """Turn a regex match into a TextMessage."""
+        author_str = REDACTED_AUTHOR_REGEX.sub('', match.group(1).strip())
+
+        # If the Sender: is redacted that means it's from self.author
+        return TextMessage(
+            author=self.author if (author_str.startswith('+') or not author_str) else author_str,
+            author_str=author_str if author_str.startswith('+') else None,  # Preserve phone numbers
+            id_confirmed=not self.is_attribution_uncertain(),
+            text=match.group(4).strip(),
+            timestamp_str=match.group(2).strip(),
+        )
+
     def _extract_timestamp(self) -> datetime:
         for match in MSG_REGEX.finditer(self.text):
             timestamp_str = match.group(2).strip()

epstein_files/documents/other_file.py

@@ -1,7 +1,7 @@
 import re
 import logging
 import warnings
-from dataclasses import dataclass
+from dataclasses import asdict, dataclass
 from datetime import datetime
 
 import datefinder
@@ -15,7 +15,7 @@ from rich.text import Text
 from epstein_files.documents.document import CLOSE_PROPERTIES_CHAR, WHITESPACE_REGEX, Document
 from epstein_files.util.constant.strings import *
 from epstein_files.util.constants import *
-from epstein_files.util.doc_cfg import FINANCIAL_REPORTS_AUTHORS, DocCfg
+from epstein_files.util.doc_cfg import FINANCIAL_REPORTS_AUTHORS, DocCfg, Metadata
 from epstein_files.util.data import escape_single_quotes, remove_timezone, uniquify
 from epstein_files.util.file_helper import FILENAME_LENGTH
 from epstein_files.util.env import args
@@ -83,11 +83,10 @@ UNINTERESTING_PREFIXES = FINANCIAL_REPORTS_AUTHORS + [
     NOBEL_CHARITABLE_TRUST,
     'Nautilus',
     'New Yorker',
-    NYT_ARTICLE,
-    NYT_COLUMN,
+    NYT,
     PALM_BEACH_CODE_ENFORCEMENT,
-    PALM_BEACH_DAILY_ARTICLE,
-    PALM_BEACH_POST_ARTICLE,
+    PALM_BEACH_DAILY_NEWS,
+    PALM_BEACH_POST,
     PALM_BEACH_TSV,
     PALM_BEACH_WATER_COMMITTEE,
     PAUL_KRASSNER,
@@ -102,6 +101,7 @@ UNINTERESTING_PREFIXES = FINANCIAL_REPORTS_AUTHORS + [
     SHIMON_POST_ARTICLE,
     SINGLE_PAGE,
     STACEY_PLASKETT,
+    'Tatler',
     TERJE_ROD_LARSEN,
     TEXT_OF_US_LAW,
     TRANSLATION,
@@ -113,7 +113,7 @@ UNINTERESTING_PREFIXES = FINANCIAL_REPORTS_AUTHORS + [
     'U.S. News',
     'US Office',
     'Vanity Fair',
-    VI_DAILY_NEWS_ARTICLE,
+    VI_DAILY_NEWS,
     WAPO,
 ]
 
@@ -127,7 +127,7 @@ class OtherFile(Document):
 
         if self.config is None and VI_DAILY_NEWS_REGEX.search(self.text):
             self.log(f"Creating synthetic config for VI Daily News article...", logging.INFO)
-            self.config = DocCfg(id=self.file_id, description=VI_DAILY_NEWS_ARTICLE, category=ARTICLE)
+            self.config = DocCfg(id=self.file_id, author=VI_DAILY_NEWS, category=ARTICLE, description='article')
 
     def category(self) -> str | None:
         return self.config and self.config.category
@@ -175,6 +175,11 @@ class OtherFile(Document):
 
         return True
 
+    def metadata(self) -> Metadata:
+        metadata = super().metadata()
+        metadata['is_interesting'] = self.is_interesting()
+        return metadata
+
     def preview_text(self) -> str:
         return WHITESPACE_REGEX.sub(' ', self.text)[0:PREVIEW_CHARS]
 

epstein_files/epstein_files.py

@@ -21,11 +21,11 @@ from epstein_files.documents.messenger_log import MSG_REGEX, MessengerLog
 from epstein_files.documents.other_file import OtherFile
 from epstein_files.util.constant.output_files import PICKLED_PATH
 from epstein_files.util.constant.strings import *
-from epstein_files.util.constant.urls import (EPSTEIN_WEB, JMAIL, epsteinify_name_url, epstein_web_person_url,
-     search_jmail_url, search_twitter_url)
+from epstein_files.util.constant.urls import (EPSTEIN_MEDIA, EPSTEIN_WEB, JMAIL, epstein_media_person_url,
+     epsteinify_name_url, epstein_web_person_url, search_jmail_url, search_twitter_url)
 from epstein_files.util.constants import *
 from epstein_files.util.data import dict_sets_to_lists, json_safe, sort_dict
-from epstein_files.util.doc_cfg import EmailCfg
+from epstein_files.util.doc_cfg import EmailCfg, Metadata
 from epstein_files.util.env import args, logger
 from epstein_files.util.file_helper import DOCS_DIR, file_size_str
 from epstein_files.util.highlighted_group import get_info_for_name, get_style_for_name
@@ -68,6 +68,7 @@ class EpsteinFiles:
         """Iterate through files and build appropriate objects."""
         self.all_files = [f for f in DOCS_DIR.iterdir() if f.is_file() and not f.name.startswith('.')]
         documents = []
+        file_type_count = defaultdict(int)
 
         # Read through and classify all the files
         for file_arg in self.all_files:
@@ -75,12 +76,13 @@ class EpsteinFiles:
             document = Document(file_arg)
 
             if document.length == 0:
-                logger.warning(f"Skipping empty file: {document}")
+                logger.warning(f"Skipping empty file: {document}]")
                 continue
 
             cls = document_cls(document)
             documents.append(cls(file_arg, text=document.text))
             logger.info(str(documents[-1]))
+            file_type_count[cls.__name__] += 1
 
             if doc_timer.seconds_since_start() > SLOW_FILE_SECONDS:
                 doc_timer.print_at_checkpoint(f"Slow file: {documents[-1]} processed")
@@ -195,15 +197,13 @@ class EpsteinFiles:
     def imessage_logs_for(self, author: str | None | list[str | None]) -> Sequence[MessengerLog]:
         return MessengerLog.logs_for(author, self.imessage_logs)
 
-    def identified_imessage_log_count(self) -> int:
-        return len([log for log in self.imessage_logs if log.author])
-
     def json_metadata(self) -> str:
+        """Create a JSON string containing metadata for all the files."""
         metadata = {
-            EMAIL_CLASS: [json_safe(d.metadata()) for d in self.emails],
-            JSON_FILE_CLASS: [json_safe(d.metadata()) for d in self.json_files],
-            MESSENGER_LOG_CLASS: [json_safe(d.metadata()) for d in self.imessage_logs],
-            OTHER_FILE_CLASS: [json_safe(d.metadata()) for d in self.other_files if not isinstance(d, JsonFile)],
+            EMAIL_CLASS: _sorted_metadata(self.emails),
+            JSON_FILE_CLASS: _sorted_metadata(self.json_files),
+            MESSENGER_LOG_CLASS: _sorted_metadata(self.imessage_logs),
+            OTHER_FILE_CLASS: _sorted_metadata(self.non_json_other_files()),
         }
 
         return json.dumps(metadata, indent=4, sort_keys=True)
@@ -216,7 +216,7 @@ class EpsteinFiles:
         add_cols_to_table(table, ['File Type', 'Files', 'Author Known', 'Author Unknown', 'Duplicates'])
 
         def add_row(label: str, docs: list):
-            known = None if isinstance(docs[0], JsonFile) else len([d for d in docs if d.author])
+            known = None if isinstance(docs[0], JsonFile) else Document.known_author_count(docs)
 
             table.add_row(
                 label,
@@ -274,7 +274,7 @@ class EpsteinFiles:
     def print_emailer_counts_table(self) -> None:
         footer = f"Identified authors of {self.attributed_email_count():,} emails out of {len(self.emails):,}."
         counts_table = Table(title=f"Email Counts", caption=footer, header_style="bold")
-        add_cols_to_table(counts_table, ['Name', 'Count', 'Sent', "Recv'd", JMAIL, EPSTEIN_WEB, 'Twitter'])
+        add_cols_to_table(counts_table, ['Name', 'Count', 'Sent', "Recv'd", JMAIL, EPSTEIN_MEDIA, EPSTEIN_WEB, 'Twitter'])
 
         emailer_counts = {
             emailer: self.email_author_counts[emailer] + self.email_recipient_counts[emailer]
@@ -290,7 +290,8 @@ class EpsteinFiles:
                 str(self.email_author_counts[p]),
                 str(self.email_recipient_counts[p]),
                 '' if p is None else link_text_obj(search_jmail_url(p), JMAIL),
-                '' if not is_ok_for_epstein_web(p) else link_text_obj(epstein_web_person_url(p), EPSTEIN_WEB.lower()),
+                '' if not is_ok_for_epstein_web(p) else link_text_obj(epstein_media_person_url(p), EPSTEIN_MEDIA),
+                '' if not is_ok_for_epstein_web(p) else link_text_obj(epstein_web_person_url(p), EPSTEIN_WEB),
                 '' if p is None else link_text_obj(search_twitter_url(p), 'search X'),
             )
 
@@ -299,7 +300,7 @@ class EpsteinFiles:
     def print_imessage_summary(self) -> None:
         """Print summary table and stats for text messages."""
         console.print(MessengerLog.summary_table(self.imessage_logs))
-        text_summary_msg = f"\nDeanonymized {self.identified_imessage_log_count()} of "
+        text_summary_msg = f"\nDeanonymized {Document.known_author_count(self.imessage_logs)} of "
         text_summary_msg += f"{len(self.imessage_logs)} {TEXT_MESSAGE} logs found in {len(self.all_files):,} files."
         console.print(text_summary_msg)
         imessage_msg_count = sum([len(log.messages()) for log in self.imessage_logs])
@@ -394,3 +395,8 @@ def is_ok_for_epstein_web(name: str | None) -> bool:
         return False
 
     return True
+
+
+def _sorted_metadata(docs: Sequence[Document]) -> list[Metadata]:
+    docs_sorted_by_id = sorted(docs, key=lambda d: d.file_id)
+    return [json_safe(d.metadata()) for d in docs_sorted_by_id]

epstein_files/util/constant/names.py

@@ -198,14 +198,10 @@ OSBORNE_LLP = f"{IAN_OSBORNE} & Partners LLP"  # Ian Osborne's PR firm
 TRUMP_ORG = 'Trump Organization'
 UBS = 'UBS'
 
-# Locations
-PALM_BEACH = 'Palm Beach'
-VIRGIN_ISLANDS = 'Virgin Islands'
-
 # First and last names that should be made part of a highlighting regex for emailers
 NAMES_TO_NOT_HIGHLIGHT: list[str] = [name.lower() for name in [
-    'Al', 'Alfredo', 'Allen', 'Alex', 'Alexander', 'Amanda', 'Andres', 'Andrew',
-    'Bard', 'Barry', 'Bill', 'Black', 'Brad', 'Bruce',
+    'Al', 'Alan', 'Alfredo', 'Allen', 'Alex', 'Alexander', 'Amanda', 'Andres', 'Andrew',
+    'Bard', 'Barry', 'Bill', 'Black', 'Boris', 'Brad', 'Bruce',
     'Carolyn', 'Chris', 'Christina',
     'Dan', 'Daniel', 'Danny', 'Darren', 'Dave', 'David',
     'Ed', 'Edward', 'Edwards', 'Epstein', 'Eric', 'Erika', 'Etienne',
@@ -215,10 +211,10 @@ NAMES_TO_NOT_HIGHLIGHT: list[str] = [name.lower() for name in [
     'Ian',
     'Jack', 'James', 'Jay', 'Jean', 'Jeff', 'Jeffrey', 'Jennifer', 'Jeremy', 'jessica', 'Joel', 'John', 'Jon', 'Jonathan', 'Joseph', 'Jr',
     'Kahn', 'Katherine', 'Ken', 'Kevin',
-    'Leon', 'Lesley', 'Linda', 'Link', 'Lisa',
+    'Larry', 'Leon', 'Lesley', 'Linda', 'Link', 'Lisa',
     'Mann', 'Marc', 'Marie', 'Mark', 'Martin', 'Melanie', 'Michael', 'Mike', 'Miller', 'Mitchell', 'Miles', 'Morris', 'Moskowitz',
     'Nancy', 'Neal', 'New',
-    'Paul', 'Paula', 'Pen', 'Peter', 'Philip',
+    'Paul', 'Paula', 'Pen', 'Peter', 'Philip', 'Prince',
     'Randall', 'Reid', 'Richard', 'Robert', 'Rodriguez', 'Roger', 'Rosenberg', 'Ross', 'Roth', 'Rubin',
     'Scott', 'Sean', 'Stanley', 'Stern', 'Stephen', 'Steve', 'Steven', 'Stone', 'Susan',
     'The', 'Thomas', 'Tim', 'Tom', 'Tyler',
@@ -228,25 +224,25 @@ NAMES_TO_NOT_HIGHLIGHT: list[str] = [name.lower() for name in [
 ]]
 
 # Names to color white in the word counts
-OTHER_NAMES = """
-    aaron albert alberto alec alex alexandra alice allen anderson andre andres ann anna anne ariana arthur
-    baldwin barack barbro barry ben benjamin berger bert binant bob bonner boyden brad bradley brady branson bruce bruno burton
-    chapman charles charlie chris christopher clint cohen colin collins conway
-    dave davis dean debra deborah dennis diana diane diaz dickinson dixon dominique don dylan
-    ed edmond elizabeth emily entwistle erik erika etienne evelyn
+OTHER_NAMES = NAMES_TO_NOT_HIGHLIGHT + """
+    aaron albert alberto alec alexandra alice anderson andre ann anna anne ariana arthur
+    baldwin barack ben benjamin berger bert binant bob bonner boyden bradley brady branson bruno bryant burton
+    chapman charles charlie christopher clint cohen colin collins conway
+    davis dean debra deborah dennis diana diane diaz dickinson dixon dominique don dylan
+    edmond elizabeth emily entwistle erik evelyn
     ferguson flachsbart francis franco frank
-    gardner gary geoff geoffrey george gilbert goldberg gonzalez gould graham greene guarino gwyneth
+    gardner gary geoff geoffrey gilbert gloria goldberg gonzalez gould graham greene guarino gwyneth
     hancock harold harrison harry helen hirsch hofstadter horowitz hussein
     isaac isaacson
-    jamie james jane janet jason jen jennifer jim joe joel johnson jones julie justin
-    kate kathy kelly kevin kim kruger kyle
-    leonard lenny lieberman louis lynch lynn
-    marcus marianne matt matthew melissa michele michelle mike mitchell moore moscowitz
+    jamie jane janet jason jen jim joe johnson jones josh julie justin
+    kate kathy kelly kim kruger kyle
+    leo leonard lenny leslie lieberman louis lynch lynn
+    marcus marianne matt matthew melissa michele michelle moore moscowitz
     nicole nussbaum
-    paul paula paulson philip philippe
-    rafael ray richardson rob robin rodriguez ron rudolph ryan
-    sara sarah seligman serge sergey silverman sloman smith snowden sorkin stanley steele stevie stewart susan
-    ted theresa thompson tiffany tim timothy tom
+    paulson philippe
+    rafael ray richardson rob robin ron rudolph ryan
+    sara sarah seligman serge sergey silverman sloman smith snowden sorkin steele stevie stewart
+    ted theresa thompson tiffany timothy
     valeria
     walter warren weinstein weiss william
     zach zack

epstein_files/util/constant/strings.py

@@ -30,6 +30,10 @@ REPUTATION = 'reputation'
 SOCIAL = 'social'
 SPEECH = 'speech'
 
+# Locations
+PALM_BEACH = 'Palm Beach'
+VIRGIN_ISLANDS = 'Virgin Islands'
+
 # Publications
 BBC = 'BBC'
 BLOOMBERG = 'Bloomberg'
@@ -38,10 +42,12 @@ DAILY_MAIL = 'Daily Mail'
 DAILY_TELEGRAPH = "Daily Telegraph"
 LA_TIMES = 'LA Times'
 MIAMI_HERALD = 'Miami Herald'
-NYT_ARTICLE = 'NYT article about'
-NYT_COLUMN = 'NYT column about'
+NYT = "New York Times"
+PALM_BEACH_DAILY_NEWS = f'{PALM_BEACH} Daily News'
+PALM_BEACH_POST = f'{PALM_BEACH} Post'
 THE_REAL_DEAL = 'The Real Deal'
 WAPO = 'WaPo'
+VI_DAILY_NEWS = f'{VIRGIN_ISLANDS} Daily News'
 
 # Site types
 EMAIL = 'email'

epstein_files/util/constant/urls.py

@@ -71,6 +71,7 @@ epsteinify_name_url = lambda name: f"{EPSTEINIFY_URL}/?name={urllib.parse.quote(
 epstein_media_doc_url = lambda file_stem: build_doc_url(DOC_LINK_BASE_URLS[EPSTEIN_MEDIA], file_stem, True)
 epstein_media_doc_link_markup = lambda filename_or_id, style = TEXT_LINK: external_doc_link_markup(EPSTEIN_MEDIA, filename_or_id, style)
 epstein_media_doc_link_txt = lambda filename_or_id, style = TEXT_LINK: Text.from_markup(epstein_media_doc_link_markup(filename_or_id, style))
+epstein_media_person_url = lambda person: f"{EPSTEIN_MEDIA_URL}/people/{parameterize(person)}"
 
 epstein_web_doc_url = lambda file_stem: f"{DOC_LINK_BASE_URLS[EPSTEIN_WEB]}/{file_stem}.jpg"
 epstein_web_person_url = lambda person: f"{EPSTEIN_WEB_URL}/{parameterize(person)}"