GitHub API v3 issue. The returned commits depend on the limiting dates.

gh_apiv3_issue.ipynb

{
 "metadata": {
  "name": "",
  "signature": "sha256:4deae5bf7a285baace185bde84b70061f6c13793432b599e2f371ce042292a58"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import os\n",
      "import json\n",
      "from datetime import datetime\n",
      "\n",
      "import requests\n",
      "import re\n",
      "\n",
      "user = \"juanlu001\"\n",
      "password = os.environ.get('GITHUB_PASSWORD')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def load_commits(since, until, author=None):\n",
      "    auth = user, password\n",
      "    if author:\n",
      "        base_url = \"https://api.github.com/repos/{}/{}/commits?since={}&until={}&author={}\".format(\n",
      "                        \"matplotlib\", \"matplotlib\", since, until, author)\n",
      "    else:\n",
      "        base_url = \"https://api.github.com/repos/{}/{}/commits?since={}&until={}\".format(\n",
      "                        \"matplotlib\", \"matplotlib\", since, until)\n",
      "    link_p = re.compile('<(.+?)>; rel=\"(first|next|prev)\"')\n",
      "    res = requests.get(base_url, auth=auth)\n",
      "    res_obj = json.loads(res.text)\n",
      "    \n",
      "    for i in range(10):\n",
      "        if 'Link' in res.headers:\n",
      "            links = res.headers['Link'].split(', ')\n",
      "            for link in links:\n",
      "                url_link, kind = link_p.match(link).groups()\n",
      "                if kind == 'next':\n",
      "                    url_next = url_link\n",
      "                    break\n",
      "            else:\n",
      "                continue\n",
      "            res = requests.get(url_next, auth=auth)\n",
      "            res_obj += json.loads(res.text)\n",
      "    return res_obj"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data1 = load_commits(\"2013-10-01T00:00:00Z\", \"2013-10-11T00:00:00Z\", \"NelleV\")\n",
      "for c in data1:\n",
      "    if c['author']:\n",
      "        if c['author']['login'] == \"NelleV\":\n",
      "            print(c['sha'][:10], c['author']['login'], c['commit']['committer']['date'], c['commit']['message'].split('\\n')[0][:60])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "26498aa247 NelleV 2013-10-09T11:24:06Z Merge pull request #2491 from pelson/pep8_re-enable\n",
        "0fab5a7eba NelleV 2013-10-08T13:57:43Z Merge pull request #2502 from pelson/gca_doc\n",
        "1129d9c6b0 NelleV 2013-10-02T08:19:46Z ENH better error message when wrong cmap name.\n"
       ]
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data2 = load_commits(\"2013-10-01T00:00:00Z\", \"2013-10-11T00:00:00Z\")\n",
      "for c in data2:\n",
      "    if c['author']:\n",
      "        if c['author']['login'] == \"NelleV\":\n",
      "            print(c['sha'][:10], c['author']['login'], c['commit']['committer']['date'], c['commit']['message'].split('\\n')[0])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "26498aa247 NelleV 2013-10-09T11:24:06Z Merge pull request #2491 from pelson/pep8_re-enable\n",
        "0fab5a7eba NelleV 2013-10-08T13:57:43Z Merge pull request #2502 from pelson/gca_doc\n",
        "1129d9c6b0 NelleV 2013-10-02T08:19:46Z ENH better error message when wrong cmap name.\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data3 = load_commits(\"2013-10-01T00:00:00Z\", \"2013-11-01T00:00:00Z\")\n",
      "for c in data3:\n",
      "    if c['author']:\n",
      "        if c['author']['login'] == \"NelleV\":\n",
      "            print(c['sha'][:10], c['author']['login'], c['commit']['committer']['date'], c['commit']['message'].split('\\n')[0])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "26498aa247 NelleV 2013-10-09T11:24:06Z Merge pull request #2491 from pelson/pep8_re-enable\n",
        "0fab5a7eba NelleV 2013-10-08T13:57:43Z Merge pull request #2502 from pelson/gca_doc\n",
        "1129d9c6b0 NelleV 2013-10-02T08:19:46Z ENH better error message when wrong cmap name.\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data4 = load_commits(\"2013-10-01T00:00:00Z\", \"2013-11-01T00:00:00Z\", \"NelleV\")\n",
      "for c in data4:\n",
      "    if c['author']:\n",
      "        if c['author']['login'] == \"NelleV\":\n",
      "            print(c['sha'][:10], c['author']['login'], c['commit']['committer']['date'], c['commit']['message'].split('\\n')[0])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "26498aa247 NelleV 2013-10-09T11:24:06Z Merge pull request #2491 from pelson/pep8_re-enable\n",
        "0fab5a7eba NelleV 2013-10-08T13:57:43Z Merge pull request #2502 from pelson/gca_doc\n",
        "1129d9c6b0 NelleV 2013-10-02T08:19:46Z ENH better error message when wrong cmap name.\n"
       ]
      }
     ],
     "prompt_number": 6
    }
   ],
   "metadata": {}
  }
 ]
}

Issue solved https://developer.github.com/changes/2014-05-09-improved-pagination-for-the-repository-commits-api/ thanks @izuzak!