##// END OF EJS Templates
Add note to docs about problems with proxy for HTML notebook....
Thomas Kluyver -
Show More
@@ -1,262 +1,273 b''
1 1 .. _htmlnotebook:
2 2
3 3 =========================
4 4 An HTML Notebook IPython
5 5 =========================
6 6
7 7 .. seealso::
8 8
9 9 :ref:`Installation requirements <installnotebook>` for the Notebook.
10 10
11 11 The IPython Notebook consists of two related components:
12 12
13 13 * An JSON based Notebook document format for recording and distributing
14 14 Python code and rich text.
15 15 * A web-based user interface for authoring and running notebook documents.
16 16
17 17 The Notebook can be used by starting the Notebook server with the
18 18 command::
19 19
20 20 $ ipython notebook
21 21
22 22 Note that by default, the notebook doesn't load pylab, it's just a normal
23 23 IPython session like any other. If you want pylab support, you must use::
24 24
25 25 $ ipython notebook --pylab
26 26
27 27 which will behave similar to the terminal and Qt console versions, using your
28 28 default matplotlib backend and providing floating interactive plot windows. If
29 29 you want inline figures, you must manually select the ``inline`` backend::
30 30
31 31 $ ipython notebook --pylab=inline
32 32
33 33 This server uses the same ZeroMQ-based two process kernel architecture as
34 34 the QT Console as well Tornado for serving HTTP/S requests. Some of the main
35 35 features of the Notebook include:
36 36
37 37 * Display rich data (png/html/latex/svg) in the browser as a result of
38 38 computations.
39 39 * Compose text cells using HTML and Markdown.
40 40 * Import and export notebook documents in range of formats (.ipynb, .py).
41 41 * In browser syntax highlighting, tab completion and autoindentation.
42 42 * Inline matplotlib plots that can be stored in Notebook documents and opened
43 43 later.
44 44
45 45 See :ref:`our installation documentation <install_index>` for directions on
46 46 how to install the notebook and its dependencies.
47 47
48 48 .. note::
49 49
50 50 You can start more than one notebook server at the same time, if you want to
51 51 work on notebooks in different directories. By default the first notebook
52 52 server starts in port 8888, later notebooks search for random ports near
53 53 that one. You can also manually specify the port with the ``--port``
54 54 option.
55 55
56 56
57 57 Basic Usage
58 58 ===========
59 59
60 60 The landing page of the notebook server application, which we call the IPython
61 61 Notebook *dashboard*, shows the notebooks currently available in the directory
62 62 in which the application was started, and allows you to create new notebooks.
63 63
64 64 A notebook is a combination of two things:
65 65
66 66 1. An interactive session connected to an IPython kernel, controlled by a web
67 67 application that can send input to the console and display many types of output
68 68 (text, graphics, mathematics and more). This is the same kernel used by the
69 69 :ref:`Qt console <qtconsole>`, but in this case the web console sends input in
70 70 persistent cells that you can edit in-place instead of the vertically scrolling
71 71 terminal style used by the Qt console.
72 72
73 73 2. A document that can save the inputs and outputs of the session as well as
74 74 additional text that accompanies the code but is not meant for execution. In
75 75 this way, notebook files serve as a complete computational record of a session
76 76 including explanatory text and mathematics, code and resulting figures. These
77 77 documents are internally JSON files and are saved with the ``.ipynb``
78 78 extension.
79 79
80 80 If you have ever used the Mathematica or Sage notebooks (the latter is also
81 81 web-based__) you should feel right at home. If you have not, you should be
82 82 able to learn how to use it in just a few minutes.
83 83
84 84 .. __: http://sagenb.org
85 85
86 86
87 87 Creating and editing notebooks
88 88 ------------------------------
89 89
90 90 You can create new notebooks from the dashboard with the ``New Notebook``
91 91 button or open existing ones by clicking on their name. Once in a notebook,
92 92 your browser tab will reflect the name of that notebook (prefixed with "IPy:").
93 93 The URL for that notebook is not meant to be human-readable and is *not*
94 94 persistent across invocations of the notebook server.
95 95
96 96 You can also drag and drop into the area listing files any python file: it
97 97 will be imported into a notebook with the same name (but ``.ipynb`` extension)
98 98 located in the directory where the notebook server was started. This notebook
99 99 will consist of a single cell with all the code in the file, which you can
100 100 later manually partition into individual cells for gradual execution, add text
101 101 and graphics, etc.
102 102
103 103 Workflow and limitations
104 104 ------------------------
105 105
106 106 The normal workflow in a notebook is quite similar to a normal IPython session,
107 107 with the difference that you can edit a cell in-place multiple times until you
108 108 obtain the desired results rather than having to rerun separate scripts with
109 109 the ``%run`` magic (though magics also work in the notebook). Typically
110 110 you'll work on a problem in pieces, organizing related pieces into cells and
111 111 moving forward as previous parts work correctly. This is much more convenient
112 112 for interactive exploration than breaking up a computation into scripts that
113 113 must be executed together, especially if parts of them take a long time to run
114 114 (In the traditional terminal-based IPython, you can use tricks with namespaces
115 115 and ``%run -i`` to achieve this capability, but we think the notebook is a more
116 116 natural solution for that kind of problem).
117 117
118 118 The only significant limitation the notebook currently has, compared to the qt
119 119 console, is that it can not run any code that expects input from the kernel
120 120 (such as scripts that call :func:`raw_input`). Very importantly, this means
121 121 that the ``%debug`` magic does *not* work in the notebook! We intend to
122 122 correct this limitation, but in the meantime, there is a way to debug problems
123 123 in the notebook: you can attach a Qt console to your existing notebook kernel,
124 124 and run ``%debug`` from the Qt console. If your notebook is running on a local
125 125 computer (i.e. if you are accessing it via your localhost address at
126 126 127.0.0.1), you can just type ``%qtconsole`` in the notebook and a Qt console
127 127 will open up connected to that same kernel.
128 128
129 129 In general, the notebook server prints the full details of how to connect to
130 130 each kernel at the terminal, with lines like::
131 131
132 132 [IPKernelApp] To connect another client to this kernel, use:
133 133 [IPKernelApp] --existing kernel-3bb93edd-6b5a-455c-99c8-3b658f45dde5.json
134 134
135 135 This is the name of a JSON file that contains all the port and validation
136 136 information necessary to connect to the kernel. You can manually start a
137 137 qt console with::
138 138
139 139 ipython qtconsole --existing kernel-3bb93edd-6b5a-455c-99c8-3b658f45dde5.json
140 140
141 141 and if you only have a single kernel running, simply typing::
142 142
143 143 ipython qtconsole --existing
144 144
145 145 will automatically find it (it will always find the most recently started
146 146 kernel if there is more than one). You can also request this connection data
147 147 by typing ``%connect_info``; this will print the same file information as well
148 148 as the content of the JSON data structure it contains.
149 149
150 150
151 151 Text input
152 152 ----------
153 153
154 154 In addition to code cells and the output they produce (such as figures), you
155 155 can also type text not meant for execution. To type text, change the type of a
156 156 cell from ``Code`` to ``Markdown`` by using the button or the :kbd:`Ctrl-m m`
157 157 keybinding (see below). You can then type any text in Markdown_ syntax, as
158 158 well as mathematical expressions if you use ``$...$`` for inline math or
159 159 ``$$...$$`` for displayed math.
160 160
161 161 Exporting a notebook
162 162 --------------------
163 163
164 164 If you want to provide others with a static HTML or PDF view of your notebook,
165 165 use the ``Print`` button. This opens a static view of the document, which you
166 166 can print to PDF using your operating system's facilities, or save to a file
167 167 with your web browser's 'Save' option (note that typically, this will create
168 168 both an html file *and* a directory called `notebook_name_files` next to it
169 169 that contains all the necessary style information, so if you intend to share
170 170 this, you must send the directory along with the main html file).
171 171
172 172 The `Download` button lets you save a notebook file to the Download area
173 173 configured by your web browser (particularly useful if you are running the
174 174 notebook server on a remote host and need a file locally). The notebook is
175 175 saved by default with the ``.ipynb`` extension and the files contain JSON data
176 176 that is not meant for human editing or consumption. But you can always export
177 177 the input part of a notebook to a plain python script by choosing Python format
178 178 in the `Download` drop list. This removes all output and saves the text cells
179 179 in comment areas.
180 180
181 181 .. warning::
182 182
183 183 While in simple cases you can roundtrip a notebook to Python, edit the
184 184 python file and import it back without loss, this is in general *not
185 185 guaranteed to work at all*. As the notebook format evolves in complexity,
186 186 there will be attributes of the notebook that will not survive a roundtrip
187 187 through the Python form. You should think of the Python format as a way to
188 188 output a script version of a notebook and the import capabilities as a way
189 189 to load existing code to get a notebook started. But the Python version is
190 190 *not* an alternate notebook format.
191 191
192 192
193 193 Keyboard use
194 194 ------------
195 195
196 196 All actions in the notebook can be achieved with the mouse, but we have also
197 197 added keyboard shortcuts for the most common ones, so that productive use of
198 198 the notebook can be achieved with minimal mouse intervention. The main
199 199 key bindings you need to remember are:
200 200
201 201 * :kbd:`Shift-Enter`: execute the current cell (similar to the Qt console),
202 202 show output (if any) and create a new cell below. Note that in the notebook,
203 203 simply using :kbd:`Enter` *never* forces execution, it simply inserts a new
204 204 line in the current cell. Therefore, in the notebook you must always use
205 205 :kbd:`Shift-Enter` to get execution (or use the mouse and click on the ``Run
206 206 Selected`` button).
207 207
208 208 * :kbd:`Ctrl-Enter`: execute the current cell in "terminal mode", where any
209 209 output is shown but the cursor stays in the current cell, whose input
210 210 area is flushed empty. This is convenient to do quick in-place experiments
211 211 or query things like filesystem content without creating additional cells you
212 212 may not want saved in your notebook.
213 213
214 214 * :kbd:`Ctrl-m`: this is the prefix for all other keybindings, which consist
215 215 of an additional single letter. Type :kbd:`Ctrl-m h` (that is, the sole
216 216 letter :kbd:`h` after :kbd:`Ctrl-m`) and IPython will show you the remaining
217 217 available keybindings.
218 218
219 219 Security
220 220 ========
221 221
222 222 You can protect your notebook server with a simple single-password by
223 223 setting the :attr:`NotebookApp.password` configurable. You can prepare a
224 224 hashed password using the function :func:`IPython.lib.security.passwd`:
225 225
226 226 .. sourcecode:: ipython
227 227
228 228 In [1]: from IPython.lib import passwd
229 229 In [2]: passwd()
230 230 Enter password:
231 231 Verify password:
232 232 Out[2]: 'sha1:67c9e60bb8b6:9ffede0825894254b2e042ea597d771089e11aed'
233 233
234 234 .. note::
235 235
236 236 :func:`~IPython.lib.security.passwd` can also take the password as a string
237 237 argument. **Do not** pass it as an argument inside an IPython session, as it
238 238 will be saved in your input history.
239 239
240 240 You can then add this to your :file:`ipython_notebook_config.py`, e.g.::
241 241
242 242 # Password to use for web authentication
243 243 c.NotebookApp.password = u'sha1:67c9e60bb8b6:9ffede0825894254b2e042ea597d771089e11aed'
244 244
245 245 When using a password, it is a good idea to also use SSL, so that your password
246 246 is not sent in the clear. You can start the notebook to communicate via a secure
247 247 protocol mode using a self-signed certificate by typing::
248 248
249 249 $ ipython notebook --certfile=mycert.pem
250 250
251 251 .. note::
252 252
253 253 A self-signed certificate can be generated with openssl. For example::
254 254
255 255 $ openssl req -x509 -nodes -days 365 -newkey rsa:1024 -keyout mycert.pem -out mycert.pem
256 256
257 Known Issues
258 ============
259
260 When behind a proxy, especially if your system or browser is set to autodetect
261 the proxy, the html notebook might fail to connect to the server's websockets,
262 and present you with a warning at startup. In this case, you need to configure
263 your system not to use the proxy for the server's address.
264
265 In Firefox, for example, go to the Preferences panel, Advanced section,
266 Network tab, click 'Settings...', and add the address of the notebook server
267 to the 'No proxy for' field.
257 268
258 269 Notebook document format
259 270 ========================
260 271
261 272
262 273 .. _Markdown: http://daringfireball.net/projects/markdown/basics
General Comments 0
You need to be logged in to leave comments. Login now