ohmyscrapper 0.2.3__tar.gz → 0.3.4__tar.gz

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.3
 Name: ohmyscrapper
-Version: 0.2.3
-Summary: This project aims to create a text-based scraper containing links to create a final PDF with general information about job openings.
+Version: 0.3.4
+Summary: OhMyScrapper scrapes texts and urls looking for links and jobs-data to create a final report with general information about job positions.
 Author: Cesar Cardoso
 Author-email: Cesar Cardoso <hello@cesarcardoso.cc>
 Requires-Dist: beautifulsoup4>=4.14.3
@@ -16,16 +16,17 @@ Requires-Dist: urlextract>=1.9.0
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown
 
-# OhMyScrapper - v0.2.3
+# 🐶 OhMyScrapper - v0.3.4
 
-This project aims to create a text-based scraper containing links to create a
-final PDF with general information about job openings.
+OhMyScrapper scrapes texts and urls looking for links and jobs-data to create a
+final report with general information about job positions.
 
 ## Scope
 
 - Read texts;
-- Extract links;
-- Use meta og:tags to extract information;
+- Extract and load urls;
+- Scrapes the urls looking for og:tags and titles;
+- Export a list of links with relevant information;
 
 ## Installation
 
@@ -50,7 +51,7 @@ uvx ohmyscrapper --version
 
 OhMyScrapper works in 3 stages:
 
-1. It collects and loads urls from a text (by default `input/_chat.txt`) in a database;
+1. It collects and loads urls from a text in a database;
 2. It scraps/access the collected urls and read what is relevant. If it finds new urls, they are collected as well;
 3. Export a list of urls in CSV files;
 
@@ -58,7 +59,7 @@ You can do 3 stages with the command:
 ```shell
 ohmyscrapper start
 ```
-> Remember to add your text file in the folder `/input` with the name `_chat.txt`!
+> Remember to add your text file in the folder `/input` with the name that finishes with `.txt`!
 
 You will find the exported files in the folder `/output` like this:
 - `/output/report.csv`
@@ -70,15 +71,14 @@ You will find the exported files in the folder `/output` like this:
 
 ### BUT: if you want to do step by step, here it is:
 
-First we load a text file you would like to look for urls, the idea here is to
-use the whatsapp history, but it works with any txt file.
+First we load a text file you would like to look for urls. It it works with any txt file.
 
-The default file is `input/_chat.txt`. If you have the default file you just use
-the command `load`:
+The default folder is `/input`. Put one or more text (finished with `.txt`) files
+in this folder and use the command `load`:
 ```shell
 ohmyscrapper load
 ```
-or, if you have another file, just use the argument `-file` like this:
+or, if you have another file in a different folder, just use the argument `-file` like this:
 ```shell
 ohmyscrapper load -file=my-text-file.txt
 ```

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/README.md

@@ -1,13 +1,14 @@
-# OhMyScrapper - v0.2.3
+# 🐶 OhMyScrapper - v0.3.4
 
-This project aims to create a text-based scraper containing links to create a
-final PDF with general information about job openings.
+OhMyScrapper scrapes texts and urls looking for links and jobs-data to create a
+final report with general information about job positions.
 
 ## Scope
 
 - Read texts;
-- Extract links;
-- Use meta og:tags to extract information;
+- Extract and load urls;
+- Scrapes the urls looking for og:tags and titles;
+- Export a list of links with relevant information;
 
 ## Installation
 
@@ -32,7 +33,7 @@ uvx ohmyscrapper --version
 
 OhMyScrapper works in 3 stages:
 
-1. It collects and loads urls from a text (by default `input/_chat.txt`) in a database;
+1. It collects and loads urls from a text in a database;
 2. It scraps/access the collected urls and read what is relevant. If it finds new urls, they are collected as well;
 3. Export a list of urls in CSV files;
 
@@ -40,7 +41,7 @@ You can do 3 stages with the command:
 ```shell
 ohmyscrapper start
 ```
-> Remember to add your text file in the folder `/input` with the name `_chat.txt`!
+> Remember to add your text file in the folder `/input` with the name that finishes with `.txt`!
 
 You will find the exported files in the folder `/output` like this:
 - `/output/report.csv`
@@ -52,15 +53,14 @@ You will find the exported files in the folder `/output` like this:
 
 ### BUT: if you want to do step by step, here it is:
 
-First we load a text file you would like to look for urls, the idea here is to
-use the whatsapp history, but it works with any txt file.
+First we load a text file you would like to look for urls. It it works with any txt file.
 
-The default file is `input/_chat.txt`. If you have the default file you just use
-the command `load`:
+The default folder is `/input`. Put one or more text (finished with `.txt`) files
+in this folder and use the command `load`:
 ```shell
 ohmyscrapper load
 ```
-or, if you have another file, just use the argument `-file` like this:
+or, if you have another file in a different folder, just use the argument `-file` like this:
 ```shell
 ohmyscrapper load -file=my-text-file.txt
 ```

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "ohmyscrapper"
-version = "0.2.3"
-description = "This project aims to create a text-based scraper containing links to create a final PDF with general information about job openings."
+version = "0.3.4"
+description = "OhMyScrapper scrapes texts and urls looking for links and jobs-data to create a final report with general information about job positions."
 readme = "README.md"
 authors = [
     { name = "Cesar Cardoso", email = "hello@cesarcardoso.cc" }

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/__init__.py

@@ -19,20 +19,22 @@ from ohmyscrapper.modules.merge_dbs import merge_dbs
 
 def main():
     parser = argparse.ArgumentParser(prog="ohmyscrapper")
-    parser.add_argument("--version", action="version", version="%(prog)s v0.2.3")
+    parser.add_argument("--version", action="version", version="%(prog)s v0.3.4")
 
     subparsers = parser.add_subparsers(dest="command", help="Available commands")
     start_parser = subparsers.add_parser(
-        "start", help="Make the entire process of loading, processing and exporting with the default configuration."
+        "start",
+        help="Make the entire process of 📦 loading, 🐶 scraping and 📜🖋️ exporting with the default configuration.",
     )
 
     start_parser.add_argument(
-        "--ai", default=False, help="Make the entire process of loading, processing, reprocessing with AI and exporting with the default configuration.", action="store_true"
+        "--ai",
+        default=False,
+        help="Make the entire process of loading, processing, reprocessing with AI and exporting with the default configuration.",
+        action="store_true",
     )
 
-    ai_process_parser = subparsers.add_parser(
-        "ai", help="Process with AI."
-    )
+    ai_process_parser = subparsers.add_parser("ai", help="Process with AI.")
     ai_process_parser.add_argument(
         "--history", default=False, help="Reprocess ai history", action="store_true"
     )
@@ -51,12 +53,13 @@ def main():
         "--recursive", default=False, help="Run in recursive mode", action="store_true"
     )
 
-    load_txt_parser = subparsers.add_parser("load", help="Load txt file")
+    load_txt_parser = subparsers.add_parser("load", help="📦 Load txt file")
+    load_txt_parser.add_argument("-file", default=None, help="File path.")
     load_txt_parser.add_argument(
-        "-file", default="input/_chat.txt", help="File path. Default is input/_chat.txt"
+        "--verbose", default=False, help="Run in verbose mode", action="store_true"
     )
 
-    scrap_urls_parser = subparsers.add_parser("scrap-urls", help="Scrap urls")
+    scrap_urls_parser = subparsers.add_parser("scrap-urls", help="🐶 Scrap urls")
     scrap_urls_parser.add_argument(
         "--recursive", default=False, help="Run in recursive mode", action="store_true"
     )
@@ -69,8 +72,11 @@ def main():
     scrap_urls_parser.add_argument(
         "--only-parents", default=False, help="Only parents urls", action="store_true"
     )
+    scrap_urls_parser.add_argument(
+        "--verbose", default=False, help="Run in verbose mode", action="store_true"
+    )
 
-    sniff_url_parser = subparsers.add_parser("sniff-url", help="Check url")
+    sniff_url_parser = subparsers.add_parser("sniff-url", help="🐕 Sniff/Check url")
     sniff_url_parser.add_argument(
         "url", default="https://cesarcardoso.cc/", help="Url to sniff"
     )
@@ -82,7 +88,7 @@ def main():
     show_urls_parser.add_argument("--limit", default=0, help="Limit of lines to show")
     show_urls_parser.add_argument("-url", default="", help="Url to show")
 
-    export_parser = subparsers.add_parser("export", help="Export urls to csv.")
+    export_parser = subparsers.add_parser("export", help="📊🖋️ Export urls to csv.")
     export_parser.add_argument("--limit", default=0, help="Limit of lines to export")
     export_parser.add_argument(
         "--file",
@@ -96,14 +102,11 @@ def main():
         action="store_true",
     )
 
-    report_parser = subparsers.add_parser("report", help="Export urls report to csv.")
+    report_parser = subparsers.add_parser(
+        "report", help="📜🖋️ Export urls report to csv."
+    )
     merge_parser = subparsers.add_parser("merge_dbs", help="Merge databases.")
 
-    # TODO: What is that?
-    # seed_parser.set_defaults(func=seed)
-    # classify_urls_parser.set_defaults(func=classify_urls)
-    # load_txt_parser.set_defaults(func=load_txt)
-
     args = parser.parse_args()
 
     if args.command == "classify-urls":
@@ -111,7 +114,7 @@ def main():
         return
 
     if args.command == "load":
-        load_txt(args.file)
+        load_txt(file_name=args.file, verbose=args.verbose)
         return
 
     if args.command == "seed":
@@ -132,6 +135,7 @@ def main():
             ignore_valid_prefix=args.ignore_type,
             randomize=args.randomize,
             only_parents=args.only_parents,
+            verbose=args.verbose,
         )
         return
 
@@ -166,7 +170,12 @@ def main():
 
     if args.command == "start":
         load_txt()
-        scrap_urls(recursive=True,ignore_valid_prefix=True,randomize=False,only_parents=False)
+        scrap_urls(
+            recursive=True,
+            ignore_valid_prefix=True,
+            randomize=False,
+            only_parents=False,
+        )
         if args.ai:
             process_with_ai()
         export_urls()

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/models/urls_manager.py

@@ -17,14 +17,21 @@ def get_db_path():
 
 
 def get_db_connection():
+    if not os.path.exists(get_db_path()):
+        create_tables(sqlite3.connect(get_db_path()))
     return sqlite3.connect(get_db_path())
 
 
-# TODO: check if it makes sense
-conn = get_db_connection()
+def use_connection(func):
+    def provide_connection(*args, **kwargs):
+        global conn
+        with get_db_connection() as conn:
+            return func(*args, **kwargs)
 
+    return provide_connection
 
-def create_tables():
+
+def create_tables(conn):
 
     c = conn.cursor()
     c.execute(
@@ -38,27 +45,19 @@ def create_tables():
         "CREATE TABLE IF NOT EXISTS urls_valid_prefix (id INTEGER PRIMARY KEY, url_prefix TEXT UNIQUE, url_type TEXT)"
     )
 
-    return pd.read_sql_query("SELECT * FROM urls LIMIT 100", conn)
-
 
-# TODO: not sure this should be something. depends on the project
 def seeds():
-    create_tables()
-
     add_urls_valid_prefix("https://%.linkedin.com/posts/%", "linkedin_post")
     add_urls_valid_prefix("https://lnkd.in/%", "linkedin_redirect")
     add_urls_valid_prefix("https://%.linkedin.com/jobs/view/%", "linkedin_job")
     add_urls_valid_prefix("https://%.linkedin.com/feed/%", "linkedin_feed")
     add_urls_valid_prefix("https://%.linkedin.com/company/%", "linkedin_company")
 
-    # add_urls_valid_prefix("%.pdf", "pdf")
-    # add_url('https://imazon.org.br/categorias/artigos-cientificos/')
-
     return True
 
 
+@use_connection
 def add_urls_valid_prefix(url_prefix, url_type):
-    conn = get_db_connection()
 
     df = pd.read_sql_query(
         f"SELECT * FROM urls_valid_prefix WHERE url_prefix = '{url_prefix}'", conn
@@ -72,6 +71,7 @@ def add_urls_valid_prefix(url_prefix, url_type):
         conn.commit()
 
 
+@use_connection
 def get_urls_valid_prefix_by_type(url_type):
     df = pd.read_sql_query(
         f"SELECT * FROM urls_valid_prefix WHERE url_type = '{url_type}'", conn
@@ -79,12 +79,14 @@ def get_urls_valid_prefix_by_type(url_type):
     return df
 
 
+@use_connection
 def get_urls_valid_prefix_by_id(id):
     df = pd.read_sql_query(f"SELECT * FROM urls_valid_prefix WHERE id = '{id}'", conn)
     return df
 
 
 # TODO: pagination required
+@use_connection
 def get_urls_valid_prefix(limit=0):
     if limit > 0:
         df = pd.read_sql_query(f"SELECT * FROM urls_valid_prefix LIMIT {limit}", conn)
@@ -94,6 +96,7 @@ def get_urls_valid_prefix(limit=0):
 
 
 # TODO: pagination required
+@use_connection
 def get_urls(limit=0):
     if limit > 0:
         df = pd.read_sql_query(
@@ -104,6 +107,7 @@ def get_urls(limit=0):
     return df
 
 
+@use_connection
 def get_urls_report():
     sql = """
     WITH parent_url AS (
@@ -138,6 +142,7 @@ def get_urls_report():
     return df
 
 
+@use_connection
 def get_url_by_url(url):
     url = clean_url(url)
     df = pd.read_sql_query(f"SELECT * FROM urls WHERE url = '{url}'", conn)
@@ -145,12 +150,14 @@ def get_url_by_url(url):
     return df
 
 
+@use_connection
 def get_url_by_id(id):
     df = pd.read_sql_query(f"SELECT * FROM urls WHERE id = '{id}'", conn)
 
     return df
 
 
+@use_connection
 def get_urls_by_url_type(url_type):
     df = pd.read_sql_query(
         f"SELECT * FROM urls WHERE history = 0 AND url_type = '{url_type}'", conn
@@ -158,6 +165,7 @@ def get_urls_by_url_type(url_type):
     return df
 
 
+@use_connection
 def get_urls_by_url_type_for_ai_process(url_type="linkedin_post", limit=10):
     df = pd.read_sql_query(
         f"SELECT * FROM urls WHERE history = 0 AND url_type = '{url_type}' AND ai_processed = 0 LIMIT {limit}",
@@ -166,6 +174,7 @@ def get_urls_by_url_type_for_ai_process(url_type="linkedin_post", limit=10):
     return df
 
 
+@use_connection
 def get_url_like_unclassified(like_condition):
     df = pd.read_sql_query(
         f"SELECT * FROM urls WHERE history = 0 AND url LIKE '{like_condition}' AND url_type IS NULL",
@@ -174,6 +183,7 @@ def get_url_like_unclassified(like_condition):
     return df
 
 
+@use_connection
 def add_url(url, h1=None, parent_url=None):
     url = clean_url(url)
     c = conn.cursor()
@@ -196,6 +206,7 @@ def add_url(url, h1=None, parent_url=None):
     return get_url_by_url(url)
 
 
+@use_connection
 def add_ai_log(instructions, response, model, prompt_file, prompt_name):
     c = conn.cursor()
 
@@ -205,10 +216,14 @@ def add_ai_log(instructions, response, model, prompt_file, prompt_name):
     )
     conn.commit()
 
+
+@use_connection
 def get_ai_log():
     df = pd.read_sql_query(f"SELECT * FROM ai_log", conn)
     return df
 
+
+@use_connection
 def set_url_destiny(url, destiny):
     url = clean_url(url)
     destiny = clean_url(destiny)
@@ -222,6 +237,7 @@ def set_url_destiny(url, destiny):
     conn.commit()
 
 
+@use_connection
 def set_url_h1(url, value):
     value = str(value).strip()
     url = clean_url(url)
@@ -230,6 +246,7 @@ def set_url_h1(url, value):
     conn.commit()
 
 
+@use_connection
 def set_url_h1_by_id(id, value):
     value = str(value).strip()
 
@@ -238,29 +255,44 @@ def set_url_h1_by_id(id, value):
     conn.commit()
 
 
+@use_connection
 def set_url_ai_processed_by_id(id, json_str):
     value = 1
     value = str(value).strip()
     c = conn.cursor()
-    c.execute("UPDATE urls SET ai_processed = ? , json_ai = ? WHERE id = ?", (value, json_str, id))
+    c.execute(
+        "UPDATE urls SET ai_processed = ? , json_ai = ? WHERE id = ?",
+        (value, json_str, id),
+    )
     conn.commit()
 
+
+@use_connection
 def set_url_empty_ai_processed_by_id(id, json_str="empty result"):
     value = 1
     value = str(value).strip()
     c = conn.cursor()
-    c.execute("UPDATE urls SET ai_processed = ? , json_ai = ? WHERE ai_processed = 0 AND id = ?", (value, json_str, id))
+    c.execute(
+        "UPDATE urls SET ai_processed = ? , json_ai = ? WHERE ai_processed = 0 AND id = ?",
+        (value, json_str, id),
+    )
     conn.commit()
 
+
+@use_connection
 def set_url_ai_processed_by_url(url, json_str):
     value = 1
     value = str(value).strip()
     url = clean_url(url)
     c = conn.cursor()
-    c.execute("UPDATE urls SET ai_processed = ?, json_ai = ? WHERE url = ?", (value, json_str, url))
+    c.execute(
+        "UPDATE urls SET ai_processed = ?, json_ai = ? WHERE url = ?",
+        (value, json_str, url),
+    )
     conn.commit()
 
 
+@use_connection
 def set_url_description(url, value):
     url = clean_url(url)
     c = conn.cursor()
@@ -268,6 +300,7 @@ def set_url_description(url, value):
     conn.commit()
 
 
+@use_connection
 def set_url_description_links(url, value):
     url = clean_url(url)
     c = conn.cursor()
@@ -275,6 +308,7 @@ def set_url_description_links(url, value):
     conn.commit()
 
 
+@use_connection
 def set_url_json(url, value):
     url = clean_url(url)
     c = conn.cursor()
@@ -282,6 +316,7 @@ def set_url_json(url, value):
     conn.commit()
 
 
+@use_connection
 def set_url_error(url, value):
     url = clean_url(url)
     c = conn.cursor()
@@ -289,6 +324,7 @@ def set_url_error(url, value):
     conn.commit()
 
 
+@use_connection
 def set_url_type_by_id(url_id, url_type):
     c = conn.cursor()
     c.execute(f"UPDATE urls SET url_type = '{url_type}' WHERE id = {url_id}")
@@ -312,6 +348,7 @@ def clean_url(url):
     return url
 
 
+@use_connection
 def get_untouched_urls(
     limit=10, randomize=True, ignore_valid_prefix=False, only_parents=True
 ):
@@ -331,6 +368,7 @@ def get_untouched_urls(
     return df
 
 
+@use_connection
 def touch_url(url):
     url = clean_url(url)
     c = conn.cursor()
@@ -338,6 +376,7 @@ def touch_url(url):
     conn.commit()
 
 
+@use_connection
 def untouch_url(url):
     url = clean_url(url)
     c = conn.cursor()
@@ -345,12 +384,14 @@ def untouch_url(url):
     conn.commit()
 
 
+@use_connection
 def untouch_all_urls():
     c = conn.cursor()
     c.execute("UPDATE urls SET last_touch = NULL WHERE history = 0")
     conn.commit()
 
 
+@use_connection
 def set_all_urls_as_history():
     c = conn.cursor()
     c.execute("UPDATE urls SET history = 1")
@@ -382,9 +423,9 @@ def merge_dbs() -> None:
                 row["description"],
                 row["json"],
             )
-        # ßmerge_url(df)
 
 
+@use_connection
 def merge_url(url, h1, last_touch, created_at, description, json):
     url = clean_url(url)
     c = conn.cursor()

ohmyscrapper-0.3.4/src/ohmyscrapper/modules/classify_urls.py

@@ -0,0 +1,27 @@
+import ohmyscrapper.models.urls_manager as urls_manager
+import pandas as pd
+import time
+
+
+def classify_urls(recursive=False):
+    urls_manager.seeds()
+    df = urls_manager.get_urls_valid_prefix()
+
+    keep_alive = True
+    while keep_alive:
+        print("#️⃣  URL Classifier woke up to classify urls!")
+        for index, row_prefix in df.iterrows():
+            df_urls = urls_manager.get_url_like_unclassified(
+                like_condition=row_prefix["url_prefix"]
+            )
+            for index, row_urls in df_urls.iterrows():
+                urls_manager.set_url_type_by_id(
+                    url_id=row_urls["id"], url_type=row_prefix["url_type"]
+                )
+
+        if not recursive:
+            print("#️⃣  URL Classifier said: I'm done! See you soon...")
+            keep_alive = False
+        else:
+            print("#️⃣  URL Classifier is taking a nap...")
+            time.sleep(10)

ohmyscrapper-0.3.4/src/ohmyscrapper/modules/load_txt.py

@@ -0,0 +1,94 @@
+import os
+from urlextract import URLExtract
+import ohmyscrapper.models.urls_manager as urls_manager
+
+
+def _increment_file_name(text_file_content, file_name):
+    print(f"reading and loading file `{file_name}`... ")
+    with open(file_name, "r") as f:
+        return text_file_content + f.read()
+
+
+def load_txt(file_name=None, verbose=False):
+    if not os.path.exists("db"):
+        os.mkdir("db")
+
+    if not os.path.exists("input"):
+        os.mkdir("input")
+
+    urls_manager.seeds()
+
+    text_file_content = ""
+    if file_name is not None:
+        print(f"📖 reading file `{file_name}`... ")
+        if not os.path.exists(file_name):
+            print(f"\n file `{file_name}` not found.")
+            return
+        text_file_content = _increment_file_name(
+            text_file_content=text_file_content, file_name=file_name
+        )
+    else:
+        print("📂 reading /input directory... ")
+        dir_files = "input"
+        text_files = os.listdir(dir_files)
+        for file in text_files:
+            if not file.endswith(".txt"):
+                text_files.remove(file)
+        if len(text_files) == 0:
+            print("No text files found in /input directory!")
+            return
+        elif len(text_files) == 1:
+            print(f"📖 reading file `{dir_files}/{text_files[0]}`... ")
+            text_file_content = _increment_file_name(
+                text_file_content=text_file_content,
+                file_name=dir_files + "/" + text_files[0],
+            )
+        else:
+            print("\nChoose a text file. Use `*` for process all and `q` to quit:")
+            for index, file in enumerate(text_files):
+                print(f"[{index}]:", dir_files + "/" + file)
+
+            # TODO: there is a better way for sure!
+            text_file_option = -1
+            while text_file_option < 0 or text_file_option >= len(text_files):
+                text_file_option = input("Enter the file number: ")
+                if text_file_option == "*":
+                    for file in text_files:
+                        text_file_content = _increment_file_name(
+                            text_file_content=text_file_content,
+                            file_name=dir_files + "/" + file,
+                        )
+                        text_file_option = 0
+                elif text_file_option == "q":
+                    return
+                elif text_file_option.isdigit():
+                    text_file_option = int(text_file_option)
+                    if text_file_option >= 0 and text_file_option < len(text_files):
+                        text_file_content = _increment_file_name(
+                            text_file_content=text_file_content,
+                            file_name=dir_files
+                            + "/"
+                            + text_files[int(text_file_option)],
+                        )
+
+    print("🔎 looking for urls...")
+    urls_found = put_urls_from_string(
+        text_to_process=text_file_content, verbose=verbose
+    )
+
+    print("--------------------")
+    print("files processed")
+    print(f"📦 {urls_found} urls were extracted and packed into the database")
+
+
+def put_urls_from_string(text_to_process, parent_url=None, verbose=False):
+    if isinstance(text_to_process, str):
+        extractor = URLExtract()
+        for url in extractor.find_urls(text_to_process):
+            urls_manager.add_url(url=url, parent_url=parent_url)
+            if verbose:
+                print(url, "added")
+
+        return len(extractor.find_urls(text_to_process))
+    else:
+        return 0

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/modules/process_with_ai.py

@@ -7,9 +7,11 @@ import time
 import os
 import yaml
 import json
+
 # TODO: !!! REFACTOR !!!
 load_dotenv()
 
+
 def reprocess_ai_history():
     df = urls_manager.get_ai_log().to_dict(orient="records")
     for row in df:
@@ -17,28 +19,34 @@ def reprocess_ai_history():
 
 
 def process_ai_response(response):
-        job_positions = xml2dict(response)
-
-        for index, xml_item_children in job_positions.items():
-            for url_child_xml in xml_item_children:
-
-                url_parent = urls_manager.get_url_by_id(url_child_xml["id"])
-                if len(url_parent) > 0:
-                    url_parent = url_parent.iloc[0]
-                h1 = url_child_xml.copy()
-                del h1["id"]
-                del h1["url"]
-                h1 = " - ".join(h1.values())
-                if url_parent["description_links"] > 1 and url_child_xml["id"] != "":
-                    print("-- child updated -- \n", url_child_xml["url"] , ":", h1)
-                    urls_manager.set_url_h1(url_child_xml["url"], h1)
-                    urls_manager.set_url_ai_processed_by_url(url_child_xml["url"], str(json.dumps(url_child_xml)))
-                    if url_parent["url"] != url_child_xml["url"]:
-                        urls_manager.set_url_ai_processed_by_url(url_parent["url"], "children-update")
-                else:
-                    print("-- parent updated -- \n", url_parent["url"], ":", h1)
-                    urls_manager.set_url_h1(url_parent["url"], h1)
-                    urls_manager.set_url_ai_processed_by_url(url_parent["url"], str(json.dumps(url_child_xml)))
+    job_positions = xml2dict(response)
+
+    for index, xml_item_children in job_positions.items():
+        for url_child_xml in xml_item_children:
+
+            url_parent = urls_manager.get_url_by_id(url_child_xml["id"])
+            if len(url_parent) > 0:
+                url_parent = url_parent.iloc[0]
+            h1 = url_child_xml.copy()
+            del h1["id"]
+            del h1["url"]
+            h1 = " - ".join(h1.values())
+            if url_parent["description_links"] > 1 and url_child_xml["id"] != "":
+                print("-- child updated -- \n", url_child_xml["url"], ":", h1)
+                urls_manager.set_url_h1(url_child_xml["url"], h1)
+                urls_manager.set_url_ai_processed_by_url(
+                    url_child_xml["url"], str(json.dumps(url_child_xml))
+                )
+                if url_parent["url"] != url_child_xml["url"]:
+                    urls_manager.set_url_ai_processed_by_url(
+                        url_parent["url"], "children-update"
+                    )
+            else:
+                print("-- parent updated -- \n", url_parent["url"], ":", h1)
+                urls_manager.set_url_h1(url_parent["url"], h1)
+                urls_manager.set_url_ai_processed_by_url(
+                    url_parent["url"], str(json.dumps(url_child_xml))
+                )
 
 
 def xml2dict(xml_string):
@@ -46,19 +54,21 @@ def xml2dict(xml_string):
 
     children_items_dict = {}
     for item in soup.find_all():
-        if(item.parent.name == "[document]"):
+        if item.parent.name == "[document]":
             children_items_dict[item.name] = []
         elif item.parent.name in children_items_dict:
             children_items_dict[item.parent.name].append(_xml_children_to_dict(item))
 
     return children_items_dict
 
+
 def _xml_children_to_dict(xml):
     item_dict = {}
     for item in xml.find_all():
         item_dict[item.name] = item.text
     return item_dict
 
+
 def process_with_ai(recursive=True, triggered_times=0):
     triggered_times = triggered_times + 1
 
@@ -91,13 +101,23 @@ def process_with_ai(recursive=True, triggered_times=0):
     print("prompt:", prompt["name"])
     print("model:", prompt["model"])
     print("description:", prompt["description"])
-    prompt["instructions"] = prompt["instructions"].replace("{ohmyscrapper_texts}", texts)
+    prompt["instructions"] = prompt["instructions"].replace(
+        "{ohmyscrapper_texts}", texts
+    )
 
     # The client gets the API key from the environment variable `GEMINI_API_KEY`.
     client = genai.Client()
-    response = client.models.generate_content(model=prompt["model"], contents=prompt["instructions"])
+    response = client.models.generate_content(
+        model=prompt["model"], contents=prompt["instructions"]
+    )
     response = str(response.text)
-    urls_manager.add_ai_log(instructions=prompt["instructions"], response=response, model=prompt["model"], prompt_name=prompt["name"], prompt_file=prompt["prompt_file"])
+    urls_manager.add_ai_log(
+        instructions=prompt["instructions"],
+        response=response,
+        model=prompt["model"],
+        prompt_name=prompt["name"],
+        prompt_file=prompt["prompt_file"],
+    )
     print(response)
     print("^^^^^^")
     process_ai_response(response=response)
@@ -114,7 +134,9 @@ def process_with_ai(recursive=True, triggered_times=0):
         if triggered_times > 5:
             print("!!! This is a break to prevent budget accident$.")
             print("You triggered", triggered_times, "times the AI processing function.")
-            print("If you are sure this is correct, you can re-call this function again.")
+            print(
+                "If you are sure this is correct, you can re-call this function again."
+            )
             print("Please, check it.")
             return
 
@@ -122,6 +144,7 @@ def process_with_ai(recursive=True, triggered_times=0):
 
     return
 
+
 def _get_prompt():
     prompts_path = "prompts"
     default_prompt = """---
@@ -135,13 +158,17 @@ Process with AI this prompt: {ohmyscrapper_texts}
         os.mkdir(prompts_path)
 
         open(f"{prompts_path}/prompt.md", "w").write(default_prompt)
-        print(f"You didn't have a prompt file. One was created in the /{prompts_path} folder. You can change it there.")
+        print(
+            f"You didn't have a prompt file. One was created in the /{prompts_path} folder. You can change it there."
+        )
         return False
 
     prompt_files = os.listdir(prompts_path)
     if len(prompt_files) == 0:
         open(f"{prompts_path}/prompt.md", "w").write(default_prompt)
-        print(f"You didn't have a prompt file. One was created in the /{prompts_path} folder. You can change it there.")
+        print(
+            f"You didn't have a prompt file. One was created in the /{prompts_path} folder. You can change it there."
+        )
         return False
     prompt = {}
     if len(prompt_files) == 1:
@@ -151,8 +178,10 @@ Process with AI this prompt: {ohmyscrapper_texts}
         prompts = {}
         for index, file in enumerate(prompt_files):
             prompts[index] = _parse_prompt(prompts_path=prompts_path, prompt_file=file)
-            print(index, ":", prompts[index]['name'])
-        input_prompt = input("Type the number of the prompt you want to use or 'q' to quit: ")
+            print(index, ":", prompts[index]["name"])
+        input_prompt = input(
+            "Type the number of the prompt you want to use or 'q' to quit: "
+        )
         if input_prompt == "q":
             return False
         try:
@@ -162,6 +191,7 @@ Process with AI this prompt: {ohmyscrapper_texts}
             prompt = _get_prompt()
     return prompt
 
+
 def _parse_prompt(prompts_path, prompt_file):
     prompt = {}
     raw_prompt = open(f"{prompts_path}/{prompt_file}", "r").read().split("---")
@@ -170,6 +200,8 @@ def _parse_prompt(prompts_path, prompt_file):
     prompt["prompt_file"] = prompt_file
 
     return prompt
+
+
 # TODO: Separate gemini from basic function
 def _process_with_gemini(model, instructions):
     response = """"""

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/modules/scrap_urls.py

@@ -7,72 +7,87 @@ import time
 import random
 
 
-def process_linkedin_redirect(url_report, url):
-    print("linkedin_redirect")
+def process_linkedin_redirect(url_report, url, verbose=False):
+    if verbose:
+        print("linkedin_redirect")
 
     if url_report["total-a-links"] < 5:
         if "first-a-link" in url_report.keys():
             url_destiny = url_report["first-a-link"]
         else:
             urls_manager.set_url_error(url=url["url"], value="error: no first-a-link")
-            print("no url for:", url["url"])
+            if verbose:
+                print("no url for:", url["url"])
             return
     else:
         if "og:url" in url_report.keys():
             url_destiny = url_report["og:url"]
         else:
             urls_manager.set_url_error(url=url["url"], value="error: no og:url")
-            print("no url for:", url["url"])
+            if verbose:
+                print("no url for:", url["url"])
             return
-
-    print(url["url"], ">>", url_destiny)
+    if verbose:
+        print(url["url"], ">>", url_destiny)
     urls_manager.add_url(url=url_destiny)
     urls_manager.set_url_destiny(url=url["url"], destiny=url_destiny)
 
 
-def process_linkedin_feed(url_report, url):
-    print("linkedin_feed")
+def process_linkedin_feed(url_report, url, verbose=False):
+    if verbose:
+        print("linkedin_feed")
 
     if "og:url" in url_report.keys():
         url_destiny = url_report["og:url"]
     else:
         urls_manager.set_url_error(url=url["url"], value="error: no og:url")
-        print("no url for:", url["url"])
+        if verbose:
+            print("no url for:", url["url"])
         return
 
-    print(url["url"], ">>", url_destiny)
+    if verbose:
+        print(url["url"], ">>", url_destiny)
     urls_manager.add_url(url=url_destiny)
     urls_manager.set_url_destiny(url=url["url"], destiny=url_destiny)
 
 
-def process_linkedin_job(url_report, url):
-    print("linkedin_job")
+def process_linkedin_job(url_report, url, verbose=False):
+    if verbose:
+        print("linkedin_job")
     changed = False
     if "h1" in url_report.keys():
-        print(url["url"], ": ", url_report["h1"])
+        if verbose:
+            print(url["url"], ": ", url_report["h1"])
         urls_manager.set_url_h1(url=url["url"], value=url_report["h1"])
         changed = True
     elif "og:title" in url_report.keys():
-        print(url["url"], ": ", url_report["og:title"])
+        if verbose:
+            print(url["url"], ": ", url_report["og:title"])
         urls_manager.set_url_h1(url=url["url"], value=url_report["og:title"])
         changed = True
 
     if "description" in url_report.keys():
-        urls_manager.set_url_description(url=url["url"], value=url_report["description"])
+        urls_manager.set_url_description(
+            url=url["url"], value=url_report["description"]
+        )
         changed = True
     elif "og:description" in url_report.keys():
-        urls_manager.set_url_description(url=url["url"], value=url_report["og:description"])
+        urls_manager.set_url_description(
+            url=url["url"], value=url_report["og:description"]
+        )
         changed = True
     if not changed:
         urls_manager.set_url_error(url=url["url"], value="error: no h1 or description")
 
 
-def process_linkedin_post(url_report, url):
-    print("linkedin_post or generic")
-    print(url["url"])
+def process_linkedin_post(url_report, url, verbose=False):
+    if verbose:
+        print("linkedin_post or generic")
+        print(url["url"])
     changed = False
     if "h1" in url_report.keys():
-        print(url["url"], ": ", url_report["h1"])
+        if verbose:
+            print(url["url"], ": ", url_report["h1"])
         urls_manager.set_url_h1(url=url["url"], value=url_report["h1"])
         changed = True
     elif "og:title" in url_report.keys():
@@ -88,52 +103,50 @@ def process_linkedin_post(url_report, url):
 
     if description is not None:
         urls_manager.set_url_description(url=url["url"], value=description)
-        description_links = load_txt.put_urls_from_string(text_to_process=description, parent_url=url["url"])
+        description_links = load_txt.put_urls_from_string(
+            text_to_process=description, parent_url=url["url"]
+        )
         urls_manager.set_url_description_links(url=url["url"], value=description_links)
 
     if not changed:
         urls_manager.set_url_error(url=url["url"], value="error: no h1 or description")
 
 
-def scrap_url(url):
-    # TODO: Use get_urls_valid_prefix_by_id()
-    df = urls_manager.get_urls_valid_prefix()
-
+def scrap_url(url, verbose=False):
     # TODO: Need to change this
 
     if url["url_type"] is None:
-        print("\n\ngeneric:", url["url"])
+        if verbose:
+            print("\n\ngeneric:", url["url"])
         url["url_type"] = "generic"
     else:
-        print("\n\n", url["url_type"] + ":", url["url"])
+        if verbose:
+            print("\n\n", url["url_type"] + ":", url["url"])
     try:
         url_report = sniff_url.get_tags(url=url["url"])
     except Exception as e:
         urls_manager.set_url_error(url=url["url"], value="error")
         urls_manager.touch_url(url=url["url"])
-        print("\n\n!!! ERROR FOR:", url["url"])
-        print(
-            "\n\n!!! you can check the URL using the command sniff-url",
-            url["url"],
-            "\n\n",
-        )
+        if verbose:
+            print("\n\n!!! ERROR FOR:", url["url"])
+            print(
+                "\n\n!!! you can check the URL using the command sniff-url",
+                url["url"],
+                "\n\n",
+            )
         return
 
-    # linkedin_redirect - linkedin (https://lnkd.in/)
     if url["url_type"] == "linkedin_redirect":
-        process_linkedin_redirect(url_report=url_report, url=url)
+        process_linkedin_redirect(url_report=url_report, url=url, verbose=verbose)
 
-    # linkedin_feed - linkedin (https://%.linkedin.com/feed/)
     if url["url_type"] == "linkedin_feed":
-        process_linkedin_feed(url_report=url_report, url=url)
+        process_linkedin_feed(url_report=url_report, url=url, verbose=verbose)
 
-    # linkedin_job - linkedin (https://www.linkedin.com/jobs/)
     if url["url_type"] == "linkedin_job":
-        process_linkedin_job(url_report=url_report, url=url)
+        process_linkedin_job(url_report=url_report, url=url, verbose=verbose)
 
-    # linkedin_job - linkedin (https://www.linkedin.com/jobs/)
     if url["url_type"] == "linkedin_post" or url["url_type"] == "generic":
-        process_linkedin_post(url_report=url_report, url=url)
+        process_linkedin_post(url_report=url_report, url=url, verbose=verbose)
 
     urls_manager.set_url_json(url=url["url"], value=url_report["json"])
     urls_manager.touch_url(url=url["url"])
@@ -144,35 +157,53 @@ def isNaN(num):
 
 
 def scrap_urls(
-    recursive=False, ignore_valid_prefix=False, randomize=False, only_parents=True
+    recursive=False,
+    ignore_valid_prefix=False,
+    randomize=False,
+    only_parents=True,
+    verbose=False,
+    n_urls=0,
 ):
+    limit = 10
     classify_urls.classify_urls()
     urls = urls_manager.get_untouched_urls(
         ignore_valid_prefix=ignore_valid_prefix,
         randomize=randomize,
         only_parents=only_parents,
+        limit=limit,
     )
     if len(urls) == 0:
-        print("no urls to scrap")
+        print("📭 no urls to scrap")
+        if n_urls > 0:
+            print(f"-- 🗃️ {n_urls} scraped urls in total...")
+            print("scrapping is over...")
         return
     for index, url in urls.iterrows():
-        scrap_url(url)
-
-        wait = random.randint(15, 20)
         wait = random.randint(1, 3)
-        print("sleeping for", wait, "seconds")
+        print(
+            "🐶 Scrapper is sleeping for", wait, "seconds before scraping next url..."
+        )
         time.sleep(wait)
 
+        print("🐕 Scrapper is sniffing the url...")
+        scrap_url(url=url, verbose=verbose)
+
+    n_urls = n_urls + len(urls)
+    print(f"-- 🗃️ {n_urls} scraped urls...")
     classify_urls.classify_urls()
     if recursive:
         wait = random.randint(5, 10)
-        print("sleeping for", wait, "seconds before next round")
+        print(
+            f"🐶 Scrapper is sleeping for {wait} seconds before next round of {limit} urls"
+        )
         time.sleep(wait)
         scrap_urls(
             recursive=recursive,
             ignore_valid_prefix=ignore_valid_prefix,
             randomize=randomize,
             only_parents=only_parents,
+            verbose=verbose,
+            n_urls=n_urls,
         )
     else:
-        print("ending...")
+        print("scrapping is over...")

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/modules/seed.py

@@ -3,5 +3,5 @@ import ohmyscrapper.models.urls_manager as urls_manager
 
 def seed():
     urls_manager.seeds()
-    print("db seeded")
+    print("🫒 db seeded")
     return

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/modules/show.py

@@ -1,10 +1,13 @@
 import ohmyscrapper.models.urls_manager as urls_manager
 import math
+import os
 from rich.console import Console
 from rich.table import Table
 
 
 def export_urls(limit=0, csv_file="output/urls.csv", simplify=False):
+    if not os.path.exists("output"):
+        os.mkdir("output")
     df = urls_manager.get_urls(limit=limit)
 
     if simplify:
@@ -12,7 +15,7 @@ def export_urls(limit=0, csv_file="output/urls.csv", simplify=False):
 
     df.to_csv(csv_file, index=False)
     print("--------------------")
-    print("Urls exported to", csv_file)
+    print("📊🖋️ Urls exported to", csv_file)
 
     df.replace(
         {
@@ -22,17 +25,19 @@ def export_urls(limit=0, csv_file="output/urls.csv", simplify=False):
         inplace=True,
     )
     df.to_html(csv_file + "-preview.html", index=False)
-    print("Urls preview exported to", csv_file + "-preview.html")
+    print("📜🖋️ Urls preview exported to", csv_file + "-preview.html")
     print("--------------------")
 
 
 def export_report(csv_file="output/report.csv"):
+    if not os.path.exists("output"):
+        os.mkdir("output")
     df = urls_manager.get_urls_report()
 
     df.to_csv(csv_file, index=False)
     _clear_file(csv_file)
     print("--------------------")
-    print("Urls report exported to", csv_file)
+    print("📊🖋️ Urls report exported to", csv_file)
 
     df.replace(
         {
@@ -44,9 +49,10 @@ def export_report(csv_file="output/report.csv"):
     df.to_html(csv_file + "-preview.html", index=False)
     _clear_file(csv_file + "-preview.html")
 
-    print("Urls report preview exported to", csv_file + "-preview.html")
+    print("📜🖋️ Urls report preview exported to", csv_file + "-preview.html")
     print("--------------------")
 
+
 # TODO: Add transformation layer
 def _clear_file(txt_tile):
     with open(txt_tile, "r") as f:
@@ -56,6 +62,7 @@ def _clear_file(txt_tile):
         with open(txt_tile, "w") as f:
             f.write(content)
 
+
 def show_urls(limit=0, jump_to_page=0):
     df = urls_manager.get_urls(limit=limit)
     df.drop(columns=["json", "description"], inplace=True)

{ohmyscrapper-0.2.3 → ohmyscrapper-0.3.4}/src/ohmyscrapper/modules/untouch_all.py

@@ -3,5 +3,5 @@ import ohmyscrapper.models.urls_manager as urls_manager
 
 def untouch_all():
     urls_manager.untouch_all_urls()
-    print("urls have been untouched")
+    print("🙌 urls have been untouched")
     return

ohmyscrapper-0.2.3/src/ohmyscrapper/modules/classify_urls.py

@@ -1,23 +0,0 @@
-import ohmyscrapper.models.urls_manager as urls_manager
-import pandas as pd
-import time
-
-
-def classify_urls(recursive=False):
-    urls_manager.seeds()
-    df = urls_manager.get_urls_valid_prefix()
-
-    keep_alive = True
-    while keep_alive:
-        print("waking up!")
-        for index, row_prefix in df.iterrows():
-            df_urls = urls_manager.get_url_like_unclassified(like_condition=row_prefix["url_prefix"])
-            for index, row_urls in df_urls.iterrows():
-                urls_manager.set_url_type_by_id(url_id =row_urls["id"], url_type=row_prefix["url_type"])
-
-        if not recursive:
-            print("ending...")
-            keep_alive = False
-        else:
-            print("sleeping...")
-            time.sleep(10)

ohmyscrapper-0.2.3/src/ohmyscrapper/modules/load_txt.py

@@ -1,32 +0,0 @@
-import os
-from urlextract import URLExtract
-import ohmyscrapper.models.urls_manager as urls_manager
-
-
-def load_txt(file_name="input/_chat.txt"):
-
-    if not os.path.exists("input"):
-        os.mkdir("input")
-
-    urls_manager.create_tables()
-    urls_manager.seeds()
-    # make it recursive for all files
-    text_file_content = open(file_name, "r").read()
-
-    put_urls_from_string(text_to_process=text_file_content)
-
-    # move_it_to_processed
-    print("--------------------")
-    print(file_name, "processed")
-
-
-def put_urls_from_string(text_to_process, parent_url=None):
-    if isinstance(text_to_process, str):
-        extractor = URLExtract()
-        for url in extractor.find_urls(text_to_process):
-            urls_manager.add_url(url=url, parent_url=parent_url)
-            print(url, "added")
-
-        return len(extractor.find_urls(text_to_process))
-    else:
-        return 0