epstein-files 1.0.13__py3-none-any.whl → 1.0.14__py3-none-any.whl

epstein_files/util/rich.py

@@ -102,8 +102,13 @@ def build_highlighter(pattern: str) -> EpsteinHighlighter:
     return TempHighlighter()
 
 
-def build_table(title: str | None, **kwargs) -> Table:
-    return Table(title=title, **{**DEFAULT_TABLE_KWARGS, **kwargs})
+def build_table(title: str | None, cols: list[str | dict] | None = None, **kwargs) -> Table:
+    table = Table(title=title, **{**DEFAULT_TABLE_KWARGS, **kwargs})
+
+    if cols:
+        add_cols_to_table(table, cols)
+
+    return table
 
 
 def join_texts(txts: list[Text], join: str = ' ', encloser: str = '', encloser_style: str = 'wheat4') -> Text:
@@ -138,6 +143,13 @@ def parenthesize(msg: str | Text, style: str = '') -> Text:
     return Text('(', style=style).append(txt).append(')')
 
 
+def print_all_files_page_link(epstein_files: 'EpsteinFiles') -> None:
+    markup_msg = link_markup(SITE_URLS[EMAIL], 'the other page', style='light_slate_grey bold')
+    txt = Text.from_markup(markup_msg).append(f' is uncurated and has all {len(epstein_files.other_files)}')
+    txt.append(f" unclassifiable files and {len(epstein_files.emails):,} emails")
+    print_centered(parenthesize(txt), style='dim')
+
+
 def print_author_header(msg: str, color: str | None, footer: str | None = None) -> None:
     txt = Text(msg, justify='center')
     color = color or 'white'
@@ -183,7 +195,20 @@ def print_header(epstein_files: 'EpsteinFiles') -> None:
 
     console.print(f"{not_optimized_msg}.\n", style='dim')
     print_page_title(width=TITLE_WIDTH)
-    print_other_site_link()
+    site_type: SiteType = EMAIL if args.all_emails else TEXT_MESSAGE
+    print_starred_header(f"This is the Epstein {site_type.title()}s site", num_spaces=4, num_stars=14)
+    other_site_type: SiteType = TEXT_MESSAGE if site_type == EMAIL else EMAIL
+    other_site_msg = "another site with" + (' all of' if other_site_type == EMAIL else '')
+    other_site_msg += f" Epstein's {other_site_type}s also generated by this code"
+
+    markup_msg = link_markup(SITE_URLS[other_site_type], other_site_msg, OTHER_SITE_LINK_STYLE)
+    print_centered(parenthesize(Text.from_markup(markup_msg)), style='bold')
+    word_count_link = link_text_obj(WORD_COUNT_URL, 'most frequently used words in the emails and texts', AUX_SITE_LINK_STYLE)
+    print_centered(parenthesize(word_count_link))
+    metadata_link = link_text_obj(JSON_METADATA_URL, 'author attribution explanations', AUX_SITE_LINK_STYLE)
+    print_centered(parenthesize(metadata_link))
+    json_link = link_text_obj(WORD_COUNT_URL, "epstein's json files", AUX_SITE_LINK_STYLE)
+    print_centered(parenthesize(json_link))
     _print_external_links()
     console.line()
     _print_abbreviations_table()
@@ -247,31 +272,6 @@ def print_numbered_list_of_emailers(_list: list[str | None], epstein_files = Non
     console.line()
 
 
-def print_other_site_link(is_header: bool = True) -> None:
-    """Print a link to the emails site if we're building text messages site and vice versa."""
-    site_type: SiteType = EMAIL if args.all_emails else TEXT_MESSAGE
-    link_style = OTHER_SITE_LINK_STYLE if is_header else 'light_slate_grey bold'
-
-    if is_header:
-        print_starred_header(f"This is the Epstein {site_type.title()}s site", num_spaces=4, num_stars=14)
-
-    other_site_type: SiteType = TEXT_MESSAGE if site_type == EMAIL else EMAIL
-    other_site_msg = "another site with" + (' all of' if other_site_type == EMAIL else '')
-    other_site_msg += f" Epstein's {other_site_type}s also generated by this code"
-    markup_msg = link_markup(SITE_URLS[other_site_type], other_site_msg, link_style)
-    print_centered(parenthesize(Text.from_markup(markup_msg)), style='bold')
-
-    if not is_header:
-        return
-
-    word_count_link = link_text_obj(WORD_COUNT_URL, 'most frequently used words in the emails and texts', AUX_SITE_LINK_STYLE)
-    print_centered(parenthesize(word_count_link))
-    metadata_link = link_text_obj(JSON_METADATA_URL, 'author attribution explanations', AUX_SITE_LINK_STYLE)
-    print_centered(parenthesize(metadata_link))
-    json_link = link_text_obj(WORD_COUNT_URL, "epstein's json files", AUX_SITE_LINK_STYLE)
-    print_centered(parenthesize(json_link))
-
-
 def print_page_title(expand: bool = True, width: int | None = None) -> None:
     title_panel = Panel(Text(PAGE_TITLE, justify='center'), expand=expand, style=TITLE_STYLE, width=width)
     console.print(Align.center(vertically_pad(title_panel)))
@@ -294,7 +294,7 @@ def print_panel(msg: str, style: str = 'black on white', padding: tuple | None =
 def print_section_header(msg: str, style: str = SECTION_HEADER_STYLE, is_centered: bool = False) -> None:
     panel = Panel(Text(msg, justify='center'), expand=True, padding=(1, 1), style=style)
     panel = Align.center(panel) if is_centered else panel
-    console.print(Padding(panel, (3, 0, 1, 0)))
+    console.print(Padding(panel, (3, 2, 1, 2)))
 
 
 def print_starred_header(msg: str, num_stars: int = 7, num_spaces: int = 2, style: str = TITLE_STYLE) -> None:

epstein_files/util/word_count.py

@@ -198,7 +198,7 @@ def write_word_counts_html() -> None:
     word_count = WordCount()
 
     # Remove dupes, junk mail, and fwded articles from emails
-    emails = [e for e in epstein_files.emails if not (e.is_duplicate() or e.is_junk_mail() or e.is_fwded_article())]
+    emails = [e for e in epstein_files.non_duplicate_emails() if not (e.is_junk_mail() or e.is_fwded_article())]
 
     for email in emails:
         if specified_names and email.author not in specified_names:

{epstein_files-1.0.13.dist-info → epstein_files-1.0.14.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: epstein-files
-Version: 1.0.13
+Version: 1.0.14
 Summary: Tools for working with the Jeffrey Epstein documents released in November 2025.
 Home-page: https://michelcrypt4d4mus.github.io/epstein_text_messages/
 License: GPL-3.0-or-later
@@ -43,11 +43,12 @@ Description-Content-Type: text/markdown
 
 
 ## Usage
-
 #### Installation
 1. Requires you have a local copy of the OCR text files from the House Oversight document release in a directory `/path/to/epstein/ocr_txt_files`. You can download those OCR text files from [the Congressional Google Drive folder](https://drive.google.com/drive/folders/1ldncvdqIf6miiskDp_EDuGSDAaI_fJx8) (make sure you grab both the `001/` and `002/` folders).
 1. Use `poetry install` for easiest time installing. `pip install epstein-files` should also work, though `pipx install epstein-files` is usually better.
 
+
+#### Command Line Tools
 You need to set the `EPSTEIN_DOCS_DIR` environment variable with the path to the folder of files you just downloaded when running. You can either create a `.env` file modeled on [`.env.example`](./.env.example) (which will set it permanently) or you can run with:
 
 ```bash
@@ -63,7 +64,7 @@ epstein_generate
 # Search for a string:
 epstein_search Bannon
 # Or a regex:
-epstein_search '\bSteve\s*Bannon\b'
+epstein_search '\bSteve\s*Bannon|Jeffrey\s*Epstein\b'
 
 # Show a file with color highlighting of keywords:
 epstein_show 030999
@@ -82,6 +83,12 @@ epstein_diff 030999 020442
 The first time you run anything it will take a few minutes to fix all the janky OCR text, attribute the redacted emails, etc. After that things will be quick.
 Run `epstein_generate --help` for command line option assistance.
 
+**Optional:** There are a handful of emails that I extracted from the legal filings they were contained in. If you want to include these files in your local analysis you'll need to copy those files from the repo into your local document directory. Something like:
+
+```bash
+cp ./emails_extracted_from_legal_filings/*.txt "$EPSTEIN_DOCS_DIR"
+```
+
 
 #### As A Library
 ```python

epstein_files-1.0.14.dist-info/RECORD

@@ -0,0 +1,33 @@
+epstein_files/__init__.py,sha256=9mClHe3YkTWHulLOZIO2uaxYcff-krpEsc0vphYlwps,4975
+epstein_files/documents/communication.py,sha256=oqNsSDWe-N0jSmchIHxpihihzIWha-foFqMwKZlxyng,2057
+epstein_files/documents/document.py,sha256=1qL1V5VrrtI7uuuLvOw-JU83khRzzEURduuoh_pfghI,17114
+epstein_files/documents/email.py,sha256=eKfDGwbP1nYVfhozM0bSVf-5T-_l_k-0D_HVrQldm38,40904
+epstein_files/documents/emails/email_header.py,sha256=wkPfSLbmzkAeQwvhf0bAeFDLPbQT-EeG0v8vNNLYktM,7502
+epstein_files/documents/imessage/text_message.py,sha256=JPHlO-5Xgt5V5E9fDsgCEr3XOp7FyzlhIHmZhFNZcmg,2806
+epstein_files/documents/json_file.py,sha256=WcZW5NNqA67rHTdopbOGtup00muNaLlvrNgKb-K4zO8,1504
+epstein_files/documents/messenger_log.py,sha256=le-cj_ZBDUb_jnxP_F05KlnPjh0nDbgpPSWl4-HL-J4,6502
+epstein_files/documents/other_file.py,sha256=tJmEH3cGeEcRIZW5F-RuSJprYM_2r1jEbsF_kjSsA3E,9788
+epstein_files/epstein_files.py,sha256=aO2vntbA7DonSk96xB5us5VaDSpp79Mpu8eYJ_g_8Qo,20025
+epstein_files/util/constant/common_words.py,sha256=aR0UjoWmxyR49XS-DtHECQ1CiA_bK8hNP6CQ1TS9yZA,3696
+epstein_files/util/constant/html.py,sha256=9U098TGzlghGg4WfxLYHyub5JGR17Dv7VP5i2MSu8Kk,1415
+epstein_files/util/constant/names.py,sha256=smC9xsR9E9WECXi_ygK_6lrcEPZWZINLZrp_-BthUtA,10377
+epstein_files/util/constant/output_files.py,sha256=BkV4_gmdj46RfGy5SFYp6dgTty3FtlBth5YGmaGutls,1700
+epstein_files/util/constant/strings.py,sha256=etCmLZPuriVYDap4iG_0Nfd92P1cvRpK2R7ZpVAGDrw,1942
+epstein_files/util/constant/urls.py,sha256=EKzpL1np3cdyYvUlgv1zl9IJU8Vrg8jXpAu6YpM0_Bk,5283
+epstein_files/util/constants.py,sha256=W-4ibPdv8I62uNyoBUR7sWWiqUWWr2n414WnZNA_Who,112605
+epstein_files/util/data.py,sha256=JccGFZGiCGm7XtwpQTocIjGYOr6hTUpEPwHhjyW9Xnc,3164
+epstein_files/util/doc_cfg.py,sha256=uVUgZF7p5QMhbPEVauQWlRWQRhP0Imr5XaNQtMk0E7c,9053
+epstein_files/util/env.py,sha256=zg-TJ8ea6nTa0PyYgY97Zksre8au1ELv2i8vOizZt7Y,5645
+epstein_files/util/file_helper.py,sha256=tacTe1GcAnckPFvjMgxRRSLnFgr2aVIYsgfDR_C9uXk,2780
+epstein_files/util/highlighted_group.py,sha256=mNcdML0h88e2voEiUJxpgAVXHh3jQ8f0XV8qVdZMkBM,36550
+epstein_files/util/logging.py,sha256=fuREq06xUUI3DfCV2JE-8QM-sQKxpLDj0_AYFO6qR1M,1983
+epstein_files/util/output.py,sha256=bb_-jRoFNNMHXPSaptEXdkbqNDJYOpNSs6dkJKvi_HQ,7581
+epstein_files/util/rich.py,sha256=bQPlOEZ6qTpVF4yPnwmlBk4K11zH5RJjwbWeBIOUUGE,15191
+epstein_files/util/search_result.py,sha256=1fxe0KPBQXBk4dLfu6m0QXIzYfZCzvaSkWqvghJGzxY,567
+epstein_files/util/timer.py,sha256=8hxW4Y1JcTUfnBrHh7sL2pM9xu1sL4HFQM4CmmzTarU,837
+epstein_files/util/word_count.py,sha256=dTBNqX98sj7W46O0uit76I4hYBhm0bsY1Q7RHwWehiU,9349
+epstein_files-1.0.14.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
+epstein_files-1.0.14.dist-info/METADATA,sha256=AUw8p6SQuvkJONsHErig0L--FyAU2y892DYpdpimqtg,5867
+epstein_files-1.0.14.dist-info/WHEEL,sha256=d2fvjOD7sXsVzChCqf0Ty0JbHKBaLYwDbGQDwQTnJ50,88
+epstein_files-1.0.14.dist-info/entry_points.txt,sha256=5qYgwAXpxegeAicD_rzda_trDRnUC51F5UVDpcZ7j6Q,240
+epstein_files-1.0.14.dist-info/RECORD,,

epstein_files-1.0.13.dist-info/RECORD

@@ -1,33 +0,0 @@
-epstein_files/__init__.py,sha256=qVFB7sS6XSlZX-ByyDwdbGSn2h06aoX2Mx8WcgRb-To,4710
-epstein_files/documents/communication.py,sha256=i6eob0ucnI7AY1sGuJlzrtU_C0mxEtrPEQZE4yNeM9w,2049
-epstein_files/documents/document.py,sha256=VsXIJhY7HTUlQrnTzYv5VLtYugdb-c2ClrfInPb5l2M,17095
-epstein_files/documents/email.py,sha256=P2c8Z23qaiWj6qWf4QKTuLvqKDoUJmu7SyzJ6FQCmPA,40727
-epstein_files/documents/emails/email_header.py,sha256=wkPfSLbmzkAeQwvhf0bAeFDLPbQT-EeG0v8vNNLYktM,7502
-epstein_files/documents/imessage/text_message.py,sha256=4gFvTfulj_Su10lNQl6Hq_p9ArTrSmn5pfC22YRJXjI,2794
-epstein_files/documents/json_file.py,sha256=tIYTwA3FYkwVZSpXvFYyUoH9m2sGYCD1U0ttamH6r1o,1306
-epstein_files/documents/messenger_log.py,sha256=BW3J52a9XxNVoK7SxtfJGydSNPoeKGE0PkdkMpf9sJ0,6294
-epstein_files/documents/other_file.py,sha256=pnl_q1o7ur3eeqGPwsYL2qbM3Y8O9LX6j6LbWnoxAiE,9939
-epstein_files/epstein_files.py,sha256=sbgS7GlrlWo1kuH4YXUSvbYX_SApgrwhpx6_qdHmrBM,19119
-epstein_files/util/constant/common_words.py,sha256=aR0UjoWmxyR49XS-DtHECQ1CiA_bK8hNP6CQ1TS9yZA,3696
-epstein_files/util/constant/html.py,sha256=9U098TGzlghGg4WfxLYHyub5JGR17Dv7VP5i2MSu8Kk,1415
-epstein_files/util/constant/names.py,sha256=ObLJCvit_C1cwLc-7Wcjh-C5oQrpww1HLk6DrEvnu8E,10368
-epstein_files/util/constant/output_files.py,sha256=BkV4_gmdj46RfGy5SFYp6dgTty3FtlBth5YGmaGutls,1700
-epstein_files/util/constant/strings.py,sha256=02DwbhAe8qBRq5HOUFx5FafXJ1P2-RJf9TCVu2b7UDQ,1932
-epstein_files/util/constant/urls.py,sha256=EKzpL1np3cdyYvUlgv1zl9IJU8Vrg8jXpAu6YpM0_Bk,5283
-epstein_files/util/constants.py,sha256=tSHrgzDOTx0KwGy4HI7ShIwaRMX1o7otW3CkKvRplZs,112121
-epstein_files/util/data.py,sha256=gOaETo4LvN3VgRvA1dB2_NRDL3wiL-xBICZ2jfD-knk,2997
-epstein_files/util/doc_cfg.py,sha256=irQljeGXlB9wbRbP05pWQZMS2-DbhdugfVGD8Q2UhSg,9006
-epstein_files/util/env.py,sha256=HnYcfHSNkwVJ_T75Woy43_OpDyxD0KHPj3GxcVx86N4,5751
-epstein_files/util/file_helper.py,sha256=tacTe1GcAnckPFvjMgxRRSLnFgr2aVIYsgfDR_C9uXk,2780
-epstein_files/util/highlighted_group.py,sha256=rKJsUQfPzQwMMu5--QWsxNFxW303RI8yq69sw4vi2RA,36720
-epstein_files/util/logging.py,sha256=fuREq06xUUI3DfCV2JE-8QM-sQKxpLDj0_AYFO6qR1M,1983
-epstein_files/util/output.py,sha256=XcflgSOlzUGj6FsFaK6j4Dljld8A0h_uVV7ERcI_EYw,8120
-epstein_files/util/rich.py,sha256=V8WDQPuCcubn46HW1cLZ6uT-CBAnBB17Hozsvv7Pd4g,14987
-epstein_files/util/search_result.py,sha256=1fxe0KPBQXBk4dLfu6m0QXIzYfZCzvaSkWqvghJGzxY,567
-epstein_files/util/timer.py,sha256=8hxW4Y1JcTUfnBrHh7sL2pM9xu1sL4HFQM4CmmzTarU,837
-epstein_files/util/word_count.py,sha256=8qBTuq3d0Q-3fwiuECKWi2RfL-KUiZD8TciwvfL0D_o,9353
-epstein_files-1.0.13.dist-info/LICENSE,sha256=OXLcl0T2SZ8Pmy2_dmlvKuetivmyPd5m1q-Gyd-zaYY,35149
-epstein_files-1.0.13.dist-info/METADATA,sha256=ZfGSbuEOmAhzDs_a6KiKjXfhLNRwcZoXklAK1KphX7I,5480
-epstein_files-1.0.13.dist-info/WHEEL,sha256=d2fvjOD7sXsVzChCqf0Ty0JbHKBaLYwDbGQDwQTnJ50,88
-epstein_files-1.0.13.dist-info/entry_points.txt,sha256=5qYgwAXpxegeAicD_rzda_trDRnUC51F5UVDpcZ7j6Q,240
-epstein_files-1.0.13.dist-info/RECORD,,