Merge pull request #1 from sopython/dev

Dev

HISTORY.rst

@@ -2,6 +2,12 @@
 History
 =======
 
+0.2.0 (2016-07-08)
+------------------
+
+* Implemented basic eridu logic to access SE.
+* Added CLI that will get content on a set schedule.
+
 0.1.0 (2016-07-07)
 ------------------
 

eridu/__init__.py

@@ -2,4 +2,7 @@
 
 __author__ = 'Keiron J. Pizzey'
 __email__ = 'kjpizzey@gmail.com'
-__version__ = '0.1.0'
+__version__ = '0.2.0'
+
+
+from eridu.core import get_post_ids, split_post_ids, get_questions, get_answers, filter_posts_by_tag

eridu/cli.py

@@ -1,13 +1,60 @@
 # -*- coding: utf-8 -*-
 
+import time
+
 import click
+import schedule
+
+from eridu.logger import logger
+from eridu.config import FILTER_TAGS, SECONDS_BETWEEN_REQUESTS
+from eridu.core import get_post_ids, split_post_ids, get_questions, get_answers, filter_posts_by_tag
 
 @click.command()
-def main(args=None):
+@click.option('--start', default=1, help="Starting page for historic results.")
+def main(start):
     """Console script for eridu"""
-    click.echo("Replace this message by putting your code into "
-                "eridu.cli.main")
-    click.echo("See click documentation at http://click.pocoo.org/")
+    logger.info('Starting main function in Eridu service.')
+
+    page = start or 0
+    params = {'page': page}
+    tags = [s.strip() for s in FILTER_TAGS.split(',')]
+
+    logger.info('Starting on page {}'.format(page))
+    logger.info('Filtering to include the following tags: {}.'.format(tags))
+
+    def run(params):
+        page = params.get('page')
+        logger.info('Getting posts for page {}.'.format(page))
+
+        post_ids = get_post_ids(page)
+        ids = split_post_ids(post_ids['items'])
+
+        questions = get_questions(ids['question_ids'])
+        questions = filter_posts_by_tag(questions['items'], tags)
+        for question in questions:
+            try:
+                print(question)
+            except UnicodeEncodeError:
+                pass
+
+        print('\n\n')
+
+        answers = get_answers(ids['answer_ids'])
+        answers = filter_posts_by_tag(answers['items'], tags)
+        for answer in answers:
+            try:
+                print(answer)
+            except UnicodeEncodeError:
+                pass
+
+        params['page'] += 1
+
+
+    schedule.every(SECONDS_BETWEEN_REQUESTS).seconds.do(run, params=params)
+
+    while True:
+        schedule.run_pending()
+        time.sleep(1)
 
 
 if __name__ == "__main__":

eridu/config.py

@@ -0,0 +1,24 @@
+import os
+
+
+ACCESS_KEY = os.environ.get('STACKEXCHANGE_REQUESTS_KEY')
+SITE = os.environ.get('ERIDU_SITE', 'stackoverflow')
+
+POST_IDS_URL = "https://api.stackexchange.com/2.2/posts"
+POST_IDS_NUMBER = os.environ.get('ERIDU_POST_IDS_NUMBER', 100)
+POST_IDS_FILTER = os.environ.get('ERIDU_POST_IDS_FILTER', '!3tz1WbZW5IHcz*twZ')
+POST_IDS_SORT = os.environ.get('ERIDU_POST_IDS_SORT', 'creation')
+POST_IDS_ORDER = os.environ.get('ERIDU_POST_IDS_ORDER', 'asc')
+
+QUESTIONS_URL = "https://api.stackexchange.com/2.2/questions/{}"
+QUESTIONS_FILTER = os.environ.get('ERIDU_QUESTIONS_FILTER', '!OfZM.T7F9gRuLlvhzHoyC1Fyd3oEOAMszsZJXvHk4mw')
+QUESTIONS_SORT = os.environ.get('ERIDU_QUESTIONS_SORT', 'creation')
+QUESTIONS_ORDER = os.environ.get('ERIDU_QUESTIONS_ORDER', 'asc')
+
+ANSWERS_URL = "https://api.stackexchange.com/2.2/answers/{}"
+ANSWERS_FILTER = os.environ.get('ERIDU_ANSWERS_FILTER', '!Fcazzsr2b3Mo6cWaRk)J*C-n25')
+ANSWERS_SORT = os.environ.get('ERIDU_ANSWERS_SORT', 'creation')
+ANSWERS_ORDER = os.environ.get('ERIDU_ANSWERS_ORDER', 'asc')
+
+FILTER_TAGS = os.environ.get('ERIDU_FILTER_TAGS', 'python,python-2.x,python-3.x')
+SECONDS_BETWEEN_REQUESTS = os.environ.get('ERIDU_SECONDS_BETWEEN_REQUESTS', 300)

eridu/core.py

@@ -0,0 +1,109 @@
+import time
+
+import requests
+
+from eridu.logger import logger
+from eridu.config import ACCESS_KEY, SITE
+from eridu.config import POST_IDS_URL, POST_IDS_FILTER, POST_IDS_NUMBER, POST_IDS_SORT, POST_IDS_ORDER
+from eridu.config import QUESTIONS_URL, QUESTIONS_FILTER, QUESTIONS_SORT, QUESTIONS_ORDER
+from eridu.config import ANSWERS_URL, ANSWERS_FILTER, ANSWERS_SORT, ANSWERS_ORDER
+
+
+
+def get_questions(question_ids, url=QUESTIONS_URL, filter=QUESTIONS_FILTER, access_key=ACCESS_KEY, site=SITE, sort=QUESTIONS_SORT, order=QUESTIONS_ORDER):
+    url = url.format(';'.join([str(i) for i in question_ids]))
+
+    payload = {
+        "pagesize": 100,
+        "key": access_key,
+        "site": site,
+        "sort": sort,
+        "order": order,
+        "filter": filter,
+    }
+
+    logger.info('Getting questions with payload: {}'.format(payload))
+
+    r = requests.get(url, params=payload)
+
+    data = r.json()
+
+    if data.get('backoff') is not None:
+        time.sleep(int(data.get('backoff')))
+
+    return data
+
+def get_answers(answer_ids, url=ANSWERS_URL, filter=ANSWERS_FILTER, access_key=ACCESS_KEY, site=SITE, sort=ANSWERS_SORT, order=ANSWERS_ORDER):
+    url = url.format(';'.join([str(i) for i in answer_ids]))
+
+    payload = {
+        "pagesize": 100,
+        "key": access_key,
+        "site": site,
+        "sort": sort,
+        "order": order,
+        "filter": filter,
+    }
+
+    logger.info('Getting answers with payload: {}'.format(payload))
+
+    r = requests.get(url, params=payload)
+
+    data = r.json()
+
+    question_ids = [answer['question_id'] for answer in data['items']]
+
+    questions = get_questions(question_ids)
+
+    tags = {question['question_id']: question['tags'] for question in questions['items']}
+
+    for answer in data['items']:
+        answer['tags'] = tags[answer['question_id']]
+
+    if data.get('backoff') is not None:
+        time.sleep(int(data.get('backoff')))
+
+    return data
+
+
+def get_post_ids(page, url=POST_IDS_URL, filter=POST_IDS_FILTER, n_posts=POST_IDS_NUMBER, access_key=ACCESS_KEY, site=SITE, sort=POST_IDS_SORT, order=POST_IDS_ORDER):
+    payload = {
+        "pagesize": n_posts,
+        "page": page,
+        "key": access_key,
+        "site": site,
+        "sort": sort,
+        "order": order,
+        "filter": filter,
+    }
+    logger.info('Getting post ids with payload: {}'.format(payload))
+
+    r = requests.get(url, params=payload)
+
+    data = r.json()
+
+    if data.get('backoff') is not None:
+        time.sleep(int(data.get('backoff')))
+
+    return data
+
+
+def split_post_ids(post_ids):
+    logger.info('Splitting post ids into question and answer ids')
+
+    question_ids, answer_ids = [], []
+
+    for item in post_ids:
+        if item['post_type'] == "question":
+            question_ids.append(item['post_id'])
+        elif item['post_type'] == 'answer':
+            answer_ids.append(item['post_id'])
+
+    return {
+        "question_ids": question_ids,
+        "answer_ids": answer_ids
+    }
+
+def filter_posts_by_tag(posts, tags):
+    tags = set(tags)
+    return [post for post in posts if set(post['tags']) & tags]

eridu/logger.py

@@ -0,0 +1,11 @@
+import logging
+
+logger = logging.getLogger('eridu')
+
+formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+
+ch = logging.StreamHandler()
+ch.setFormatter(formatter)
+
+logger.setLevel(logging.INFO)
+logger.addHandler(ch)

setup.py

@@ -20,7 +20,7 @@
 
 setup(
     name='sopython-eridu',
-    version='0.1.0',
+    version='0.2.0',
     description="Project Cradle library for accessing historic SO content.",
     long_description=readme + '\n\n' + history,
     author="Keiron J. Pizzey",