From 0f0aad6662906c46d13121ac52972d0effa3d99b Mon Sep 17 00:00:00 2001
From: Yajo <Yajo@users.noreply.github.com>
Date: Mon, 4 Apr 2016 16:58:49 +0200
Subject: [PATCH] Excerpt generator.

* [8.0][html_text] Excerpt generator.

This module adds a technical utility to allow the developer to get an excerpt from any HTML chunk.

You can choose the maximum amount of words or characters if you want.

See the README and inline docstrings for further details.
---
 html_text/README.rst                    |  82 ++++++++++++++++++++++++
 html_text/__init__.py                   |   5 ++
 html_text/__openerp__.py                |  23 +++++++
 html_text/models/__init__.py            |   5 ++
 html_text/models/ir_fields_converter.py |  72 +++++++++++++++++++++
 html_text/static/description/icon.png   | Bin 0 -> 9455 bytes
 html_text/tests/__init__.py             |   5 ++
 html_text/tests/test_extractor.py       |  53 +++++++++++++++
 8 files changed, 245 insertions(+)
 create mode 100644 html_text/README.rst
 create mode 100644 html_text/__init__.py
 create mode 100644 html_text/__openerp__.py
 create mode 100644 html_text/models/__init__.py
 create mode 100644 html_text/models/ir_fields_converter.py
 create mode 100644 html_text/static/description/icon.png
 create mode 100644 html_text/tests/__init__.py
 create mode 100644 html_text/tests/test_extractor.py

diff --git a/html_text/README.rst b/html_text/README.rst
new file mode 100644
index 000000000..38eac30af
--- /dev/null
+++ b/html_text/README.rst
@@ -0,0 +1,82 @@
+.. image:: https://img.shields.io/badge/licence-AGPL--3-blue.svg
+   :target: http://www.gnu.org/licenses/agpl-3.0-standalone.html
+   :alt: License: AGPL-3
+
+====================
+Text from HTML field
+====================
+
+This module provides some technical features that allow to extract text from
+any chunk of HTML, without HTML tags or attributes. You can chose either:
+
+* To truncate the result by amount of words or characters.
+* To append an ellipsis (or any character(s)) at the end of the result.
+
+It can be used to easily generate excerpts.
+
+Usage
+=====
+
+This module just adds a technical utility, but nothing for the end user.
+
+If you are a developer and need this utility for your module, see these
+examples and read the docs inside the code.
+
+Python example::
+
+    @api.multi
+    def some_method(self):
+        # Get truncated text from an HTML field. It will 40 words and 100
+        # characters at most, and will have "..." appended at the end if it
+        # gets truncated.
+        truncated_text = self.env["ir.fields.converter"].text_from_html(
+            self.html_field, 40, 100, "...")
+
+QWeb example::
+
+    <t t-esc="env['ir.fields.converter'].text_from_html(doc.html_field)"/>
+
+.. image:: https://odoo-community.org/website/image/ir.attachment/5784_f2813bd/datas
+   :alt: Try me on Runbot
+   :target: https://runbot.odoo-community.org/runbot/149/8.0
+
+Known issues / Roadmap
+======================
+
+* An option could be added to try to respect the basic HTML tags inside the
+  excerpt (``<b>``, ``<i>``, ``<p>``, etc.).
+
+Bug Tracker
+===========
+
+Bugs are tracked on `GitHub Issues
+<https://github.com/OCA/server-tools/issues>`_. In case of trouble, please
+check there if your issue has already been reported. If you spotted it first,
+help us smashing it by providing a detailed and welcomed `feedback
+<https://github.com/OCA/
+server-tools/issues/new?body=module:%20
+html_text%0Aversion:%20
+8.0%0A%0A**Steps%20to%20reproduce**%0A-%20...%0A%0A**Current%20behavior**%0A%0A**Expected%20behavior**>`_.
+
+Credits
+=======
+
+Contributors
+------------
+
+* Jairo Llopis <yajo.sk8@gmail.com>
+
+Maintainer
+----------
+
+.. image:: https://odoo-community.org/logo.png
+   :alt: Odoo Community Association
+   :target: https://odoo-community.org
+
+This module is maintained by the OCA.
+
+OCA, or the Odoo Community Association, is a nonprofit organization whose
+mission is to support the collaborative development of Odoo features and
+promote its widespread use.
+
+To contribute to this module, please visit https://odoo-community.org.
diff --git a/html_text/__init__.py b/html_text/__init__.py
new file mode 100644
index 000000000..09356048f
--- /dev/null
+++ b/html_text/__init__.py
@@ -0,0 +1,5 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+
+from . import models
diff --git a/html_text/__openerp__.py b/html_text/__openerp__.py
new file mode 100644
index 000000000..07f931ee8
--- /dev/null
+++ b/html_text/__openerp__.py
@@ -0,0 +1,23 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+{
+    "name": "Text from HTML field",
+    "summary": "Generate excerpts from any HTML field",
+    "version": "8.0.1.0.0",
+    "category": "Tools",
+    "website": "https://grupoesoc.es",
+    "author": "Grupo ESOC Ingeniería de Servicios, "
+              "Odoo Community Association (OCA)",
+    "license": "AGPL-3",
+    "application": False,
+    "installable": True,
+    "external_dependencies": {
+        "python": [
+            "lxml.html",
+        ],
+    },
+    "depends": [
+        "base",
+    ],
+}
diff --git a/html_text/models/__init__.py b/html_text/models/__init__.py
new file mode 100644
index 000000000..11b8b13d5
--- /dev/null
+++ b/html_text/models/__init__.py
@@ -0,0 +1,5 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+
+from . import ir_fields_converter
diff --git a/html_text/models/ir_fields_converter.py b/html_text/models/ir_fields_converter.py
new file mode 100644
index 000000000..503b2c7c7
--- /dev/null
+++ b/html_text/models/ir_fields_converter.py
@@ -0,0 +1,72 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+
+import logging
+from lxml import etree, html
+from openerp import api, models
+
+_logger = logging.getLogger(__name__)
+
+
+class IrFieldsConverter(models.Model):
+    _inherit = "ir.fields.converter"
+
+    @api.model
+    def text_from_html(self, html_content, max_words=None, max_chars=None,
+                       ellipsis=u"…", fail=False):
+        """Extract text from an HTML field in a generator.
+
+        :param str html_content:
+            HTML contents from where to extract the text.
+
+        :param int max_words:
+            Maximum amount of words allowed in the resulting string.
+
+        :param int max_chars:
+            Maximum amount of characters allowed in the resulting string. If
+            you apply this limit, beware that the last word could get cut in an
+            unexpected place.
+
+        :param str ellipsis:
+            Character(s) to be appended to the end of the resulting string if
+            it gets truncated after applying limits set in :param:`max_words`
+            or :param:`max_chars`. If you want nothing applied, just set an
+            empty string.
+
+        :param bool fail:
+            If ``True``, exceptions will be raised. Otherwise, an empty string
+            will be returned on failure.
+        """
+        # Parse HTML
+        try:
+            doc = html.fromstring(html_content)
+        except (etree.XMLSyntaxError, etree.ParserError):
+            if fail:
+                raise
+            else:
+                _logger.exception("Failure parsing this HTML:\n%s",
+                                  html_content)
+                return ""
+
+        # Get words
+        words = u"".join(doc.xpath("//text()")).split()
+
+        # Truncate words
+        suffix = max_words and len(words) > max_words
+        if max_words:
+            words = words[:max_words]
+
+        # Get text
+        text = u" ".join(words)
+
+        # Truncate text
+        suffix = suffix or max_chars and len(text) > max_chars
+        if max_chars:
+            text = text[:max_chars - (len(ellipsis) if suffix else 0)].strip()
+
+        # Append ellipsis if needed
+        if suffix:
+            text += ellipsis
+
+        return text
diff --git a/html_text/static/description/icon.png b/html_text/static/description/icon.png
new file mode 100644
index 0000000000000000000000000000000000000000..3a0328b516c4980e8e44cdb63fd945757ddd132d
GIT binary patch
literal 9455
zcmW++2RxMjAAjx~&dlBk9S+%}OXg)AGE&Cb*&}<C%<R2Kc9faym6aW`f0Dh5$js*d
z_}}Z!;XIG;_cPz`_vag-p{7Ve$Uq1H00~A(?iu(VaQlMefnU3&OozZXm@69d91cGG
z;O61r&je0NdamH#&)mKsXk?Zb_)B^>d0jUxM@u(PQx^-s)6<jB#=*|j%+$$(&(Xyy
zYgd8+09XKwoa}S2?48%%ZU#L~n^g{fE40h%YEx5by;GuJ(K|vI1uO;0m}=J7R4@Bs
z>97TX<v{QF=s?r`z`m$a0ZvrhBU7}{15RN9M`j!kv_Mp+3~{||YNuFzHNvhMYm3=Q
zo!q+OJ5(%-oNM^I^M%*luKMiVL`lo`bcKGymX7i3MIFWj1i|9+CGNvn`cu-RsK0Qh
zoYlwB>`ehR4?GS^qbkof1cslKgk<Uw6DeIxZyT_?=OwX|lwC*A?ac*gHF7jmS080$
z>U)h65qZ9Oc=ml_0temigYLJfnz{IDzUf>bGs4N!v3=Z3jMq&A#7%rM5eQ#dc?k~!
zVpnB`o+K7|Al`Q_U<UrcmRnh6jExs}l(hZs0%T~Dnpu-NENdhioRv(T{Qmv>;eD$B
zfJtP*jH`siUq~{KE)`jP2|#TUEFGRryE2`i0**z#*^6~AI|YzIWy$Cu#CSLW3q=GA
z6`?GZymC;dCPk~rBS%eCb`5OLr;RUZ;D`}um=H)BfVIq%7VhiMr)_#G0N#zrNH|__
zc+blN2UAB0=617@>_<D4$IPL<k1zq(*VmlDPer&h1n6`AG;9A!_W=N$JthhQWXZ<i
zGURc6f<i*joK3xSWaOOXxAc8ES>u;MPHN;P;N#YoE=)R#i$k_`UAA>WWCcEVMh~L_
zj--gtp&|K1#58Yz*AHCTMziU1Jzt_jG0I@qAOHsk$2}yTmVkBp_eHuY$A9)>P6o~I
z%aQ?!(GqeQ-Y+b0I(m9pwgi(IIZZzsbMv+9w{PFtd_<_(LA~0H(xz<Z(Qt1jC2cC|
z6WbMo9YgON{L#ZDl$sV4*<CP(>{=FhLB@(1&qHA5EJw1>>=%q2f&^X>IQ{!GJ4e9U
z&KlB)z(84HmNgm2hg2C0>WM{E(DdPr+EeU_N@57;PC2&DmGFW_9kP&%?X4}+xWi)(
z;)z%wI5>D4a*5XwD)P--sPkoY(a~WBw;E~AW`Yue4kFa^LM3X`8x|}ZUeMnqr}>kH
zG%WWW>3ml$Yez?i%)2pbKPI7?5o?hydokgQyZsNEr{a|mLdt;X2TX(#B1j35xPnPW
z*bMSSOauW>o;*=kO8ojw91VX!qoOQb)zHJ!odWB}d+*K?#sY_jqPdg{Sm2HdYzdEx
zOGVPhVRTGPtv0o}RfVP;Nd(|CB)<HrC1(%ZOEd8PI?r9b_$Cp-${bh59Z_R7n&YCp
zl8lfMpes*8{FVm;oJH@0LLoWmxD59u?JwHz2iRrAaE0`Hc&g;t5~$P*kdeOZn9OJ3
zRczo@ZkWU)RG+hcfH~{49Vvb3D(W0wRX#|$cA0Iqy^VR~(4hqA2AFI-rK!FM!#fJ_
zGFI@iqJOPXZ!=VjTS3dMuu~9xU3K1&?th;$)cqM#WGxbDEyB$y`#9j%>I;*t&QO8h
zFfekr30S!-LHmV_Su-W+rEwYXJ^;6&3|L$mMC8*bQptyOo9;>Qb9Q9`ySe3%V$A*9
zeKEe+b0{#KWGp$F+tga)0RtI)nhMa-K@JS}2krK~n8vJ=Ngm?R!9G<~RyuU0d?nz#
z-5EK$o(!F?hmX*2Yt6+coY`6jGbb7t<dboft~zeDZwK=+l2bFWs5^+|r{J6GO9Cwl
z&SW58BRs?XdFLuhtzl7WLbQ#~z#`jAB3AbSUg6jW6@qVd2Q~9qN(izT1y*>F#6nHA
zuKk=GGJ;ZwON1iAfG$E#Y7MnZVmrY|j0eVI(DN_MNFJmyZ|;w4tf@=CCDZ#5N_0K=
z$;R~bbk?}TpfDjfB&aiQ$VA}s?P}xPERJG{kxk5~R`iRS(SK5d+Xs9swCozZISbnS
zk!)I0>t=A<-^z(cmSFz3=jZ23u13X><0b)P)^1T_))Kr`e!-pb#q&J*Q`p+B6la%C
zuVl&0duN<;uOsB3%T9Fp8t{ED108<?)`y_~Hnd9AUX7h-H?jVuU|}My+C=TjH(jKz
zqMVr0re3S$H@t{zI95qa)+Crz*5Zj}Ao%4Z><+W(nOZd?gDnfNBC3>M8WE61$So|P
zVvqH0SNtDTcs<xiU1;=a39$d&&l5EwoIH1db#`S9Kw!YC1jhR)VbCWMptlUUkpfif
zMzi-i8)WL?|C$*Q?iqbS{w<lA9XN(rD)VS<zn>UdzaMDpT=Ty0pDHHNL@Z0w$Y`XO
z2M-_r1S+GaH%pz#Uy0*w$Vdl=X=rQXEzO}d6J^R6zjM1u&c9vYLvLp?W7w(?np9x1
zE_0JSAJCPB%i7p*Wvg)pn5T`8k3-uR?*NT|J`eS#_#54p>!p(mLDvmc-3o0mX*mp_
zN*AeS<>#^-{S%W<*mz^!X$w_2dHWpcJ6^j64qFBft-o}o_Vx80o0>}Du;>kLts;$8
zC`7q$QI(dKYG`Wa8#wl@V4jVWBRGQ@1dr-hstpQL)Tl+aqVpGpbSfN>5i&QMXfiZ>
zaA?T1VGe?rpQ@;+pkrVdd{klI&jVS@I5_iz!=UMpTsa~mBga?1r}a<Xwa$pLotZk<
zsykXwQO37I=aXew7x=}YWiW)^p)|C#g+)an!@j?EcY8C0t)BlK#y{YLtkJ6=D6H-5
zp2*ANf@>RBm1WS;TT*s0f0lY=JBl66Upy)-k4J}lh=P^8(SXk~0xW=T9v*B|gzIhN
z>qsO7dFd~mgxAy4V?&)=5ieYq?zi?ZEoj)&2o)RLy=<bK!vY7J*RP!&i_xU}i*o6o
z?xN*1<rEe1L2HBkCSgiYM3a|4w?Bo7CJL7?Eh;9An1m$1t?h1v92<Xg2(#)bjbL|o
zH_H0}!Og=1dOBynXHu>@hbCRcfT5ji<pmK_U*~T(A$I-*rM$8-BCv{=@=7F)2pdtU
z5==tp!}A*&Xgf{F>gwtQGE{L*8<@Yd{zg;CsL5mvzfDY}P-wos_6PfprFVaeqNE%h
zKZhLtcQld;ZD+>=nqN~>GvROfueSzJD&<KAVm#0W>BE*}XfU|H&(FssBqY=hPCt`d
zH?@s2>I(|;fcW&YM6#V<T#ysvE$@4oRO>#!kUIP8$Nkdh0A(bEVj``-AAyYgwY~jB
zT|I7Bf@%;7aL7Wf4dZ%VqF$eiaC38OV6oy3Z#TER2G+fOCd9Iaoy6aLYbPTN{XRPz
z;U!V|vBf%H!}52L2gH_+j;`bTcQRXB+y9onc^wLm5wi3-Be}U>k_u>2Eg$=k!(l@I
zcCg+flakT2Nej3i0yn+g+}%NYb?ta;R<sv-KjYb}UVFjITQ)VAEWu*)Lz7*-f^A*H
z<25#Ynt~-Qh?$wWx4$1=T2`j@bKp!)3IXh(LC@)%WGW$+j(tGz(6#bGw&K0j=Np@B
zt}@t$R`z(>?(g5SnwsQ49U8Wng8d|{B+lyRcEDvR3+`O{zfmrmvFrL6acVP%yG98X
zo&+VBg@px@i)%o?dG(`T;n*$S5*rnyiR#=wW}}GsAcfyQpE|>a{=$Hjg=-*_K;UtD
z<sX7(ZJc+Q;#xP8uk`jnF@a^|TWw*55if6@@}%7lOBaGo9Ia-e?`RPQc^w^EWfhe}
zHWHTvDA+r}Xf5Yqpr;QU-88%QC9F_>#z-)AXwSRY?<M%E84!g*1GC?E>OPefw^iI+
z)AXz#PfEjlwTes|_{sB?4(O@fg0AJ^g8gP}ex9Ucf*@_^J(s_5jJV}c)s$`Myn|Kd
z$<h)Fm>6>}#q^n{4vN@+Os$m7KV+`}c%4)4pv@06af4-x5#wj!KKb%caK{A&Y#Rfs
z-po?Dcb1({W=6FKIUirH&(yg=*6aLCekcKwyfK^JN5{wcA3nhO(o}SK#!CINhI`-I
z1)6&n7O&ZmyFMuNwvEic#IiOAwNkR=u5it{B9n2sAJV5pNhar=j5`*N!Na;c7g!l$
z3aYBqUkqqTJ=Re-;)s!EOeij=7SQZ3Hq}ZRds%IM*PtM$wV<GYik+VfZene%bm(n6
z`r@rc^TVw7EN{|O7rORMlw)zt(Z#enc3joE#IIk!M)L8gk^go9E9AxiP9o#OqmvV>
z@;rlc*NRK7i3y5BETSKuumEN`Xu_8<BQ)z0!-JuC8x}?$qo8SEko}oUWNI(4h*VHO
zAi!Egy!f(o9aHs2dhSE6ki(be*&w&+WLOB<(b3T_Hide(NvVvLQV{BN|2?UJFaY*@
z9CXA5B_*8i5Bf6sn~d`R>GP1Ri=OK<SGIb#BCTo3qI#UBUg+dkR+2ilUwIg%XFV;l
z+>Q$@I^ko8>H6)4rjiG5{VBM>B|%`&&s^)jS|-_95&yc=GqjNo{zFkw%%HHhS~e=s
zD#sfS+-?*t|J!+ozP6KvtOl!R)@@-z24}`9{QaVLD^9VCSR2b`b!KC#o;Ki<+wXB6
zx3&O0LOWcg4&rv4QG0)4yb}7BFSEg~=IR5<g6yi=XozU}zReXL{rA%c`$IQAs<ObZ
zep*ITZ0C(~W*`?XH^l1t7&+qigAfY9tVJ5DH!~jY>#ZRj8kg}dS7_V&^%#Do==#`u
zpy6{ox?jWuR(;pg+f@mT>#HGWHAJRRDDDv~@(IDw&R>9643kK<aUZ^OhBu;1e6t#{
zxKehVgz`HT;A`FMivG<tq0OcrXwHUX_<$j<uk%m>#HN`!1vBJHnC+RM&yIh8{gG2q
zA%e*U3|N0XSRa~oX-3EAneep)@{h2vvd3Xvy$7og(sayr@95+e6~Xvi1tUqnIxoIH
zVWo*OwYElb#uyW{Imam6f2<eMTQj#)z8+N&ZY?tSPo%&2eVNU=4=?rlO<*9Twaxco
zEVE=Jh?_x>rGbjR!Y3`#gPqkv57dB6K^wRGxc9B(t|aYDGS=m$&S!NmCtrMMaUg(c
zc2qC=2Z`EEFMW-me5B)24AqF*bV5Dr-M5ig(l-WPS%CgaPzs6p_gnCIvTJ=Y<6!gT
zVt@AfYCzjjsMEGi=rDQHo0yc;HqoRNnNFeWZgcm?f;cp(6CNylj36DoL(?TS7eU#+
z7&mfr#y))+CJOXQKUMZ7QIdS9@#-}7y2K1{8)cCt0~-X0O!O?Qx#E4Og+;A2SjalQ
zs7r?qn0H044=sDN$SRG$arw~n=+T_DNdSrarmu)V6@|?1-ZB#hRn`uilTGPJ@fqEy
zGt(f0B+^JDP&f=r{#Y_wi#AVDf-y!RIXU^0jXsFpf>=Ji*TeqSY!H~AMbJdCGLhC)
zn7Rx+sXw6uYj;WRYrLd^5IZq@6JI1C^YkgnedZEYy<&4(z%Q$5yv#B<pkvWZft^C&
z;+zNo+VJNluxaDF!`gW+F0=MxsCQ~~#CYKa;ULfj5hTa8a6;d*(<eeQ7-ZQgz2et^
zf|7URfj;x*#HiF0wuBCOTEpbeD&kkENqolCm2#cQGHCeEC<&I3j+P6?sX?BPp4~46
zx-S~EJ>oo{AH8n$<d4loB?vL3RqXwK_COrQ6xRoWGOdFnWy(hhzrG8k;2k}8Zj&>a
zhb4Y3PWdr269&?V%uI$xMcUrMzl=;w<_nm*qr=c3Rl@i5wWB;e-`t7D&c-mcQl7x!
zZWB`UGcw=Y2=}~wzrfLx=uet<;m3~=8I~ZRuzvMQUQdr+yTV|ATf1Uuomr__nDf=X
zZ3WYJtHp_ri(}SQAPjv+Y+0=<GD??Na(2AG`s>fH4krOP@S&=zZ-t1jW1o@}z;xk8
z(Nz1co&El^HK^NrhVHa-_;&88vTU>_J33=%{if;BEY*J#1n59=07jrGQ#IP>@u#3A
z;!q+E1Rj3ZJ+!4bq9F8PXJ@yMgZL;>&gYA0%_Kbi8?S=XGM~dnQZQ!yBSgcZhY96H
zrWnU;k)qy`rX&&xlDyA%(a1Hhi5CWkmg(`Gb%m(HKi-7Z!LKGRP_B8@`7&hdDy5n=
z`OIxqxiVfX@OX1p(mQu>0Ai*v_cTMiw4qRt3~N<X{IYiJ3k=4u-u*nJEBnJ<OdI8P
zXmK`OYkO5a{YDhI3PL|H4m=p>Bvr9oBy0)r>w3p~V0SCm=An6@3n)>@z!|o-$HvDK
z|3D2ZMJkLE5loMKl6R^ez@Zz%S$&mbeoqH5`Bb){Ei21q&VP)hWS2tjShfFtGE+$z
zzCR$P#uktu+#!w)cX!<osOkId_HzAT-HD2N`M+v2l+zwdW%GgZbQMuhfE*hHjKXKg
z45hphqVETPDbX6wpTlZqza0gFaRGh`3L~0id)F6#%@9;w(e%PjzuD7@inuToA!B?F
zCMLJc!u{3N)s?lB-!11!GxXsCak8zQomO)gmn02f-5~OkMYnm~#jVwwYNw@LAv!KN
z-n`q1|AXw*TW>lWN1XU%K-r=s{|j?)Akf@q#3b#{6cZCuJ~gCxuMXRmI$nGtnH+-h
z+GEi!*X=AP<|fG`1>MBdTb?28JYc=fGvAi2I<$B(rs$;eoJCyR6_bc~p!XR@O-+sD
z=eH`-ye})I5ic1eL~TDmtfJ|8`0VJ*Yr=hNCd)G1p2MMz4C3^Mj?7;!w|Ly%JqmuW
zlIEW^Ft%z?*|fpXda>Jr^1noFZEwFgVV%|*XhH@acv8rdGxeEX{M$(vG{Zw+x(ei@
zmfXb22}8-?Fi`vo-YVrTH*C?a8%M=Hv9MqVH7H^J$KsD?>!SFZ;ZsvnHr_gn=7acz
z#W?0eCdVhVMWN12VV^$>WlQ?f;P^{(&pYTops|btm6aj>_Uz+hqpGwB)vWp0Cf5y<
zft8-je~nn?W11plq}N)4A{l8I7$!ks_x$PXW-2XaRFswX_BnF{R#6YIwMhAgd5F9X
zGmwdadS6(a^fjHtXg8=l?Rc0Sm%hk6E9!5cLVloEy4eh(=FwgP`)~I^5~pBEWo+F6
zSf2ncyMurJN91#cJTy_u8Y}@%!bq1RkGC~-bV@SXRd4F{R-*V<h953|jk-C&|3)z%
ze&_30-6q1)a0(!xXqA+_t(WC`HA_yYaW@>`bS+6;W5vZ(&+I<9$;-V|eNfLa5n-6%
z2(}&uGRF;p9<Q%jdNy1%e7XTzyu7EEQT(5LrnvX~T%K!IuDyHY`ZneVJu#k_1T)xA
z*yxB?y4!pOJ$DTZzBm|8OIQq+AtV25aJ+X5EJjAu><tfuiNClI2BSoMgqLnU-5t95
zEnZ(^g~1RgD=UMB({c;$HujG&%DqGF;5jI~roRT+(rOoS$6f6SsURIuAVkAeIVc~{
z5ZxLN%m%Z*Sg;qYCf3<$dE6~&w_!EBx%yl4YC~LH{FCFNRBc_I>2eS*sE*o<YSPrx
z{M(reV{~jKue#Y6ZOnqJia{fQc!UxV4m)l389OmzR6A3|2!i<P{r82jKw+zq4%@ny
znopi6g!1Vx9Bml#a~KdL2lp2C)qSTKIh43vgycQHfQb_I!kQY&p;X@Bz|{^SXsW1K
zP&Ag1CW_r6u5-4=s(W>R$@pexaqr*meB)VhmIg@h{uzkk$9~qh#cHhw#>O%)b@+(|
z^IQgqzuj~Sk(J;swEM-3TrJAPCq9k^^^`q{IItKBRXYe}e0Tdr=Huf7da3$l4<V=<
zfr<*%iOB1EY}w3tF2Cwl7a2OS&~Y-lu<s)T^9=OFU8(CeN|63BiMxf*)5gesGU<eZ
z9DigzL3+quZ1o2T<KAZbCGJx&lrl*e#}Dpv24bZO$B<Yoc5hbeP0y?@P%|Tvw||e%
zV#e^q0D2R_Oq_c+=x5vP&hg4k+df{o7$aNZCM>PdpwWDop%^}n;dD#K4s#DYA8SHZ
z&1!riV4W4R7R#C))JH1~axJ)RYnM$$lIR%6fIVA@zV{XVyx}C+a-Dt8Y9M)^KU0+H
zR4IUb2CJ{Hg>CuaXtD50jB(_Tcx=Z$^W<wsNa}nUDpNk$q{>Yu2u5kubqmwp%drJ6
z?Fo40g!Qd<-l=TQxqHEOuPX0;^z7iX?Ke^a%XT<13TA^5`4Xcw6D@Ur&VT&CUe0d}
z1GjOVF1^L@>O)l@?bD~$wzgf(nxX1OGD8fEV?TdJcZc2KoUe|oP1#=$$7ee|xbY)A
zDZq+cuTpc(fFdj^=!;{k03C69lMQ(|>uhRfRu%+!k&<F<Z7l=VnwB^RA&^APMi=Tn
zNc}%IJL~xB4OP6bJNwAw)~Ma2=UP0dhr%X=kco(it+@F<#$xrIJ8@|{Buh<)h`xg?
z_U}pe=3#zmDfdqE4`Hyn<!?&CfCp#GTeXo8;AYd1Opd&cXER8cBSOF3iFJ#XPgQVp
zZSiQf7V^Dt?ITs8aMF~e1hq0P@^oB)#byh|6q_7F1B%ST%WL~J?ySn>YOi-3|1QKB
z<?_cUy)V3go%%^l)lRa=dmY_XQG3Z{Q?52d$qG}vIe|EZbYEtrR$rh(iS)O#dU-(>
z?n?eq1XP>p-IM$Z^C;2L3itnbJZAip*Zo0aw2bs8@(s^~*8T9go!%dHcAz2lM;`yp
zD=7&xjFV$S&5uDaiScyD?B-i1ze`+CoRtz`Wn+Zl&#<i<Zx(`=7k~{%`w&<EbG}U<
z0%E^8lyw>s4&}MO{@N!ufrzjG$B79)Y2d3tBk&)TxUTw@<TSeYlROCt(gU?O((-qu
zs>QS0TEL_?njX|<LXnSC4TlId(D4W|GQ?L@$3Ue@N8p=l9-sC<=z(lPk;^sT(v2*k
zc~vp#Hp`mXjzhml2NMCh^h5)sqsan+jJ_l<a4w|G&ac02hrz8#6|kFraA~rta0}!q
zB4BE{QWY7MtxHn_xB);Avf#Lf<GG<A?Q3JVyc1p8^H}$8(Gn=_&F1!m8$sKyeue+L
z5&392wm+wO;_oXoTEJ=wxVXk}dt<d~CgUUMW_PPTe(c<7n18#mVaX)vK}-PzZq7<b
zNJbVTFaq(8ZLx|A*G$H3ugT2aVziDEGa66FVt@hr1|D{RWN6yuqlglM!rp^YG;UpG
zrnm?ek079l3VoOU^p%f^A9Yzn8IVX?^rB4LbgJ|PONhzM_0{P?S(V$OI=u5IBjfT#
z0ntLLnhg5$0fAHJaG6HCx4X7;RmvMtE}8C>@vq?Uz(nBFK5Pq7*xj#u*R&i|?7+6#
z+|r_n#SW&LXhtheZdah{ZVoqwyT{D>MC3nkFF#N)xLi{p7J1jXlmVeb;cP5?e(=f#
zuT7fv<Q3o5LspCx-RPkzLw{VsDE>jSbjS781v?7{)-X3*?>tq?)Yd)~|1{BDS(pqC
zC}~H#WXlkUW*H5CDOo<)#x7%RY)A;ShGhI5s*#cRDA8YgqG(HeKDx+#(ZQ?386dv!
zlXCO)w91~Vw4AmOcATuV653fa9R$fyK8<JV*@W33SMHM;w!OmUXar{O;_8j>ul%rG
z-<zwGD)!Y{+(T{%ob~79zpXX%zulyiy1_UHWvu@YqxKAK3e9GO6Os4R0Naujn>wfS
zihugoZyr38Im?Zuh6@RcF~t1anQu7>#lPpb#}4cOA!EM11`%f*07RqOVkmX{p~KJ9
z^zP;K#|)$`^Rb{rnH<AdQ^(;gQMcF>GH{~>1(fawV0*Z#)}M`m8-?ZJV<+e}s9wE#
z)l&az?w^5{)`S(%MRzxdNqrs1n*-=jS^_jqE*5XDrA0+VE`5^*p3CuM<&dZEeCjoz
zR;uu_H9ZPZV|fQq`Cyw4nscrVwi!fE6ciMmX$!_hN7uF;jjKG)d2@aC4ropY)8<!P
zAOHj?oPbjQ%hh|vE_1IMB)43eQpeLi&)R>etW=xJvni)8eHi`H$%#zn^WJ<U7gogJ
z?A^!2J~rI78JH|Ml2EldmKY5K8;Zx(FGcBdM<5oe#G+nd6dObqz@Af%%M*aBE_pn8
zXQo2`Bw*IwvP3#1Ev<%YocpBodO9+Rw~`5*CY3{L;qf1PxV!r%NI+#Q1f8GU7$~#U
zAA9$4&g-k=8BYi*3i@0^UX}nTQVyOf)8T(}G^Ti?Vqvk~bJRR#EAQ?u`dClPxk@{;
zxvO?%jSX`2{8|^z0*C5DR1Z^>5NLc-rqk|u&&4Z6fD_m&JfSI1Bvb?b<*n&sfl0^t
z=HnmRl`XrFvMKB%9}>PaA`m-fK6a0(8=qPkWS5bb4=v?XcWi&hRY?O5HdulRi4?fN
zlsJ*N-0Qw+Yic@s0(2uy%F@ib;GjXt01F<S%GT3P{Ck&Y*^gWuie`o_g+ZNDNIV|F
z)zEe|Ij*4$H1(<RLz0($;AD3VsUJwI@liw^?fh(V?VC`Sz7h`*Q<VYlhbHJ?&h+!W
zAJC)U;Lx_QRaSNFYbyi|7+MeNTgA+Znh}qFzf>mx5XbRo6+n|pP(&nodMoap^z{~q
ziEeaUT@Mxe3vJSfI6?uLND(CNr=#^W<1b}jzW58bIfyWTDle$mmS(|x-0|2UlX+9k
zQ<WB5+u#`K#~J$81)#?BuTOKLk|+S>^EX7Nw}?EzVoBfT(-LT|=9N@^hcn-_p&sqG
z&*oVs2JSU+N4ZD`FhCAWaS;>|wH2G*Id|?pa#@>tyxX`+4HyIArWDvVrX)2WAOQff
z0qyHu&-S@i^MS-+j--!pr4fPBj~_8({~e1bfcl0wI1kaoN>mJL6KUPQm5N7lB(ui1
zE-o%kq)&djzWJ}ob<-GfDlkB;F31j-VHKvQUGQ3sp`CwyGJk_i!y^sD0fqC@$9|jO
zOqN!r!8-p==F@ZVP=U$qSpY(gQ0)59P1&t@y?5rvg<}E+GB}2<F#{*|k0E{$&_`~)
zkzDcsi#!7r<a8lPUCMj?{CK;e|9#-xj>6NYPp4f2YFQrQtot5mn3wu_qprZ=>Ig-$
zbW26Ws~IgY>}^5w`vTB(G`PTZaDiGBo5o(tp)qli|NeV(<Rzgqw(Ze!7hEGKKx((>
z@H_=R8V39rt5J5YB2Ky?4eJJ#b`_iBe2ot~6%7mLt5t8Vwi^Jy7|jWXqa3amOIoRb
zOr}WVFP--DsS`1WpN%~)t3R!arKF^Q$e12KEqU36AWwnCBICpH4XCsfnyrHr>$I$4
z!DpKX$OKLWarN7nv@!uIA+~RNO)l$$w}p(;b>mx8pwYvu;dD_unryX_N<mojSrG!`
zgkqx4t<XLL6ZUppvjeV9PJ6dG>hT8*Tj>BTrTTL&!?O+%Rv;b?B??gSzdp?6Uug9{
zd@V08Z$BdI?fpoCS$)t4mg4rT8Q_I}h`0d-vYZ^|dOB*Q^S|xqTV*<bTMtKO;_Z)R
zRSOJrd5TFO0aO%#%-sNa{`SiQ^$$1^@oUd&^lB{MKZ>vIg?@fVFSmMpaw0qtTRbx}
z({Pg?#{2`sc9)M5N$*N|4;^t$+Q<Wh^yJ?FzJ*$wiB{j;CMy+?m1MbsJo0ubR==f9
z>P?#mo<zt1E6=Ilm*nbmYmpxfAj7*m*Wh@=7|;!%(-pviW`nuCktLveet9^$*y^>v
zGVC@I*lBVrOU-%2y!7%)fAKjpEFsgQc4{amtiHb95KQEwvf<(3T<9-Zm$xIew#P22
zc2Ix|App^>v6(3L_MCU0d3W##AB<Fxl*nmny6_RsCu%1a)so}}uCXSzwY70Y@uTxK
z=5nu(N*2HDbrIcL)t_*{*84mvmV_Y9<vxHJJ;0gUdNjyW)jDb|@|j*qR8;gsMa5Ir
z02fHv=%xyN9VLv_ZLL4y|F*p$S^^T47m{aok5{rmM{$s7^Xu2!_fo1$IG6kkG_Tgx
zFfjPm3;g>0M~3D00EWoKZqsJYT(#@w$Y_H7G22M~ApVFTRHMI_3be)Lkn#0F*V8Pq
zc}`Cjy$bE;FJ6H7p=0y#R>`}-m4(0F>%@P|?7fx{=R^uFdISRnZ2W_xQhD{YuR3t<
z{6yxu=4~JkeA;|(J6_nv#>Nvs&FuLA&PW^he@t(UwFFE8)|a!R{`E`K`i^ZnyE4$k
z;(749Ix|oi$c3QbEJ3b~D_kQsPz~fIUKym($a_7dJ?o+40*OLl^{=&oq$<#Q(yyrp
z{J-FAniyAw9tPbe&IhQ|a`DqFTVQGQ&Gq3!C2==4x{6EJwiPZ8zub-iXoUtkJiG{}
zPaR&}_fn8_z~(=;5lD-aPWD3z8PZS@AaUiomF!G8I}Mf>e~0g#BelA-5#`cj;O5>N
Xviia!U7SGha1wx#SCgwmn*{w2TRX*I

literal 0
HcmV?d00001

diff --git a/html_text/tests/__init__.py b/html_text/tests/__init__.py
new file mode 100644
index 000000000..60346a281
--- /dev/null
+++ b/html_text/tests/__init__.py
@@ -0,0 +1,5 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+
+from . import test_extractor
diff --git a/html_text/tests/test_extractor.py b/html_text/tests/test_extractor.py
new file mode 100644
index 000000000..f7d0843d3
--- /dev/null
+++ b/html_text/tests/test_extractor.py
@@ -0,0 +1,53 @@
+# -*- coding: utf-8 -*-
+# © 2016 Grupo ESOC Ingeniería de Servicios, S.L.U. - Jairo Llopis
+# License AGPL-3.0 or later (http://www.gnu.org/licenses/agpl.html).
+
+from lxml import etree
+from openerp.tests.common import TransactionCase
+
+
+class ExtractorCase(TransactionCase):
+    def setUp(self):
+        super(ExtractorCase, self).setUp()
+
+        # Shortcut
+        self.text_from_html = self.env["ir.fields.converter"].text_from_html
+
+    def test_excerpts(self):
+        """Text gets correctly extracted."""
+        html = u"""
+            <html>
+                <body>
+                    <div class="this should not appear">
+                        <h1>I'm a title</h1>
+                        <p>I'm a paragraph</p>
+                        <small>¡Pues yo soy español!</small>
+                    </div>
+                </body>
+            </html>
+            """
+
+        self.assertEqual(
+            self.text_from_html(html),
+            u"I'm a title I'm a paragraph ¡Pues yo soy español!")
+        self.assertEqual(
+            self.text_from_html(html, 8),
+            u"I'm a title I'm a paragraph ¡Pues yo…")
+        self.assertEqual(
+            self.text_from_html(html, 8, 31),
+            u"I'm a title I'm a paragraph ¡P…")
+        self.assertEqual(
+            self.text_from_html(html, 7, ellipsis=""),
+            u"I'm a title I'm a paragraph ¡Pues")
+
+    def test_empty_html(self):
+        """Empty HTML handled correctly."""
+        self.assertEqual(self.text_from_html(""), "")
+        with self.assertRaises(etree.XMLSyntaxError):
+            self.text_from_html("", fail=True)
+
+    def test_bad_html(self):
+        """Bad HTML handled correctly."""
+        self.assertEqual(self.text_from_html("<<bad>"), "")
+        with self.assertRaises(etree.ParserError):
+            self.text_from_html("<<bad>", fail=True)