From 4678c8a2a4c5f2984b2a8b3051b0376cf0c2bec4 Mon Sep 17 00:00:00 2001
From: AbSadiki <sadiki.abdeladim@gmail.com>
Date: Fri, 3 Jan 2025 16:29:26 -0500
Subject: [PATCH 1/8] fix(transcription): IS_AUDIO_TRANSCRIPTION_CAPABLE should
 be iniztialized (#194)

---
 src/markitdown/_markitdown.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index 789c1e5..6df13e3 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -33,6 +33,7 @@
 from charset_normalizer import from_path
 
 # Optional Transcription support
+IS_AUDIO_TRANSCRIPTION_CAPABLE = False
 try:
     # Using warnings' catch_warnings to catch
     # pydub's warning of ffmpeg or avconv missing

From d248621ba4e7f4f91dba22c000a17c62b394d0c1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Murat=20Can=20Kurtulu=C5=9F?= <kmt1bu@bosch.com>
Date: Sat, 4 Jan 2025 00:34:39 +0300
Subject: [PATCH 2/8] feat: outlook ".msg" file converter (#196)

* feat: outlook .msg converter
* add test, adjust docstring
---
 pyproject.toml                        |   1 +
 src/markitdown/_markitdown.py         |  75 ++++++++++++++++++++++++++
 tests/test_files/test_outlook_msg.msg | Bin 0 -> 13312 bytes
 tests/test_markitdown.py              |  13 +++++
 4 files changed, 89 insertions(+)
 create mode 100644 tests/test_files/test_outlook_msg.msg

diff --git a/pyproject.toml b/pyproject.toml
index 3e14cec..67f6825 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -35,6 +35,7 @@ dependencies = [
   "pdfminer.six",
   "puremagic",
   "pydub",
+  "olefile",
   "youtube-transcript-api",
   "SpeechRecognition",
   "pathvalidate",
diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index 6df13e3..d209b5e 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -21,6 +21,7 @@
 
 import mammoth
 import markdownify
+import olefile
 import pandas as pd
 import pdfminer
 import pdfminer.high_level
@@ -1077,6 +1078,79 @@ def _get_llm_description(self, local_path, extension, client, model, prompt=None
         return response.choices[0].message.content
 
 
+class OutlookMsgConverter(DocumentConverter):
+    """Converts Outlook .msg files to markdown by extracting email metadata and content.
+
+    Uses the olefile package to parse the .msg file structure and extract:
+    - Email headers (From, To, Subject)
+    - Email body content
+    """
+
+    def convert(
+        self, local_path: str, **kwargs: Any
+    ) -> Union[None, DocumentConverterResult]:
+        # Bail if not a MSG file
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".msg":
+            return None
+
+        try:
+            msg = olefile.OleFileIO(local_path)
+            # Extract email metadata
+            md_content = "# Email Message\n\n"
+
+            # Get headers
+            headers = {
+                "From": self._get_stream_data(msg, "__substg1.0_0C1F001F"),
+                "To": self._get_stream_data(msg, "__substg1.0_0E04001F"),
+                "Subject": self._get_stream_data(msg, "__substg1.0_0037001F"),
+            }
+
+            # Add headers to markdown
+            for key, value in headers.items():
+                if value:
+                    md_content += f"**{key}:** {value}\n"
+
+            md_content += "\n## Content\n\n"
+
+            # Get email body
+            body = self._get_stream_data(msg, "__substg1.0_1000001F")
+            if body:
+                md_content += body
+
+            msg.close()
+
+            return DocumentConverterResult(
+                title=headers.get("Subject"), text_content=md_content.strip()
+            )
+
+        except Exception as e:
+            raise FileConversionException(
+                f"Could not convert MSG file '{local_path}': {str(e)}"
+            )
+
+    def _get_stream_data(
+        self, msg: olefile.OleFileIO, stream_path: str
+    ) -> Union[str, None]:
+        """Helper to safely extract and decode stream data from the MSG file."""
+        try:
+            if msg.exists(stream_path):
+                data = msg.openstream(stream_path).read()
+                # Try UTF-16 first (common for .msg files)
+                try:
+                    return data.decode("utf-16-le").strip()
+                except UnicodeDecodeError:
+                    # Fall back to UTF-8
+                    try:
+                        return data.decode("utf-8").strip()
+                    except UnicodeDecodeError:
+                        # Last resort - ignore errors
+                        return data.decode("utf-8", errors="ignore").strip()
+        except Exception:
+            pass
+        return None
+
+
 class ZipConverter(DocumentConverter):
     """Converts ZIP files to markdown by extracting and converting all contained files.
 
@@ -1286,6 +1360,7 @@ def __init__(
         self.register_page_converter(IpynbConverter())
         self.register_page_converter(PdfConverter())
         self.register_page_converter(ZipConverter())
+        self.register_page_converter(OutlookMsgConverter())
 
     def convert(
         self, source: Union[str, requests.Response, Path], **kwargs: Any
diff --git a/tests/test_files/test_outlook_msg.msg b/tests/test_files/test_outlook_msg.msg
new file mode 100644
index 0000000000000000000000000000000000000000..05b087b77c785c8b57a479485b9715fb7dfecbb4
GIT binary patch
literal 13312
zcmeHN-BVk~6+c3bjg7ILwn=KIj;=}Trgp#*2rveM2nZVkG6mtd9nCNjlCWAKQAxyM
zJj|q-$xPdqzW6l{{rCs^(npW~gZ8mAopjpA=DE{8CG_{(s}+Li<6J|AWX6m4-g|b>
z?(dwl=j@mK*T1~{&)@y&(!b<Dx*~r0V)wM1tSk57yUWrSM9$!wWnb*>?y~s1$oPDb
zzCjk~blees#sL(WabNo9xsPR^kLX*voQG{cD~qxqeG-$RR3zgSUgBs|MoUMcvLQ*y
zNgm$|rnC%ty-lCX3-QHU3oA>L@u|sJ-`vVld}V%RIdXepa(2FN>fS;-fhfBpQ|37*
zTT+57TaaN(R(+0)K_-?ZQM!g_0enB-$5oaHWVDj^fvX7Wop!Lb`seGv)*P0aMUG0Z
z+=rz~uw@Ps6yz4P3PIYSbYq^FHX3A1=`!Rm$lIvz$Df0`45TI%L=KyFA#nD~0G>ho
zIdUD(0rSn?_K!|4B$zfmkHL<Ua$5#v6Y{(-KY=u1^gbsyA=R)vka76QAl}RJF8T~h
zO76e|*1^3bKbBv}BiQbajG(_wwA%n*ucT4)2>q_e5c+?J??|1!8`+*RQp=S5%;;7z
z(rG-6EoWdwjv*~LE)&3wqpTM?OhD=hP|J{f9a3$Z_cAPVAKz(6ejggAVa@j>3(j%$
z@cbS`S>lky%CQ9>%pQ{*Q^uzl-vvm~3%Rz<2vIhAa2};Olq603+`yY9^v8nnyd1nC
zBctF>p(pxZ+VF0}Mm@%_=w}E2Mo^N1b=OSI5Ik}O?S}CjLfIf%(nidOOTWAx_~X+*
zT)6m2K|b}Jk9S~I{gBdgUh2b>TBnUR5j;PD<Qvd6k8d0A0-sobRTd$L*OrHqz194s
z>DkPm@#u2AZ#f#9jxI%_4<;8D=cAgrll|}MtwH-T&ta{#*S>`DSTkOj!x#Ou`DpFQ
z5eM&K)}J_Lq#Sy1s?Q4OOx4fRh!O|>p2gFaTX`LD*;RuvL@Dfg=f$Hx_K7erMCT`|
z)#F5|k_{!2g>ue3);8Be{e%69Y^k{Mu(FX4If-m4m(G=)L^_kr1|KJ~#X_Z$%Adhk
zFp(`k%9YE`WGR=-kM8WCaIKgxmQbH4WzxaKdMT4B1QV(JRwg*Q(|snfwH8WjCQIqW
zW<HrKu+K^|H82ze414PjccYJM2|`zr>mx(XSUUH<lg^dKd-9c1&pSJ&3Nky7?N<<=
zwH8C+h8*#S&;b7`4>L}xnBI1ZS@T`Vlq*hVBbm!P8<}!Bxt^)(Yq-DG7iF#yL8tr7
zJ~2bVurqc8g4T&w$6c=x6h&9g&d#1odHVXgteKvKu47&=e)4-m-hKMFWY4Fabyc1|
z{Z(%JwvGMrn%UpN#c?7M3CJZwf2+nnRNMc$^}~j~TCI;@wJ<yEySM**H~x2Dhi}x-
zS4@AOoeE$DVm&5(lLA=dN&mD)XAFI_M(-H9vnptZ@8VhVe}wE8_IJt9Z&dj^e3uP<
zxmN#%q3xO#kbt3sRsMi<oZByt{{Vf$&_h-Jh@3RERQ2=ADMMR(b;;{)y}_sbE*tu4
zwSAWa_VN4WKj$qzRrJgM3{7^|zH))});=A+(}w;v0A2kD4gDK7K#ItzAo)`tJb6m{
zN2GHf-QnvpG>)^Ty<fV+<ljDb!r=Mk9}}ei{ke<pmsFNC^<%<$_(MB(Zi(mTZ#}m-
z(4f)&L9lO(Vdoo@d)N)e4z$=KbzFP=2ftQgbFq8x#cmxe_e%Ta72}7rCC9nL_g^#a
zCk*}BsSe++q2I33;4$L{(xia+3~k2(eq0+@G6clGuilRPY1GkgOefFrfE+Kz&)Co%
zKaQQ3@w0JsT;qpW{z}Hr?z8%E2GQ>e4@Tqj_2R{i4~uY6VlUX_Jv4rG{xL_0rRdb`
z;%qb)_j-z;&$|9=*R5vr&lAYZ`NGaWd-IJu|JZf0+5BVc?fmm*wLjY8I1=w}8)Z(2
zww!wf-us?K?nXX`d>)x==F7<33w;~;JIJpfv)yZW^FG-2(dg||>wEB~9lnqJ1LUj7
zKSXAmAK`ti`c1kQ-+jn8khy0GA@?H>Acv6$ksD!+qI|4L-Mc8Cs;is9_w72mzCT{*
zaW!QbUf=k%g!3V9PI>>yGiwQFMlb1AdC3P;n8euW>UJaIncE3X*)Tn$!yv8Vk9RaY
zf#z__v1)E=_U;kgJ4Bba(eo|w8GOv+wLNqrzGh{_FPpnMp2F9wcb2QwyCB|4hLGuD
zjng&kRH!`W1M}NQW3DfkGIp3+<MYl_htGWy<uIMLd2pQUL8&!<SdIUZ2mJcjX@*a&
zL$JG6f9lTEMs-t757p=&!mTgw1Vb%;7O~^c`BIBdU(t;J0PY&K_|!*>Pha6&%hb&N
z#2vs{RBM0kqqX?-l`8~CYK=dw)}L#X7GL9k)IV(ePdENtAGP?Ls~#XTwbFkGtC@EE
zxi)L@=___0(h7eVzk_J;xi8S-(^pP12#}iNPa*Dj1FxI^xF^x#zY5-u2#(YmUmyS6
z4{7l)g7<p_M{12v`|I@Qe4@p_1YX_;F}22zs>h%EKrR03;B7zua{Oo1_}qhP@o9VR
zA(@)%Pu;ox=;EK-87;nU{vp2mo2O>|;oke`@wxu%^yj``i%;L<UY)6x{X-t+AI7v6
zpLe6});~s_{nOZ==N&D60KDzye-~dr|M9G)#pl|~`%<Q6_9y;;hxLbZnil^p@cxS6
zNX_tF&)4riIHzgxuY<RK|FwxzPtp9I%>RACU+>D0i@zARKZgXX`+v@9T72GNF1p7g
zqyOFwQDf9qoQ|z8+Bdo|=~ngp^Cl>}rc|p`uV(fSV(y*6E<wBgFs8NobML`DC{t^E
z&i}ggk1?&q9|kYSfT<b2oq1_*P5d*awfLI(-^JIjKaL0do8aYsiK(^z)(TCJ@wl6D
z&%c~gj$VI#{P%dk9|14-piI>NDB_>GGr7ICe)7kX`uWd${PWh!O*Z~Hr`a{l)@k0&
z#@}oF^WILY|2TM$I{vSEz~}#tJL>r7eWh0auOj}rrfTsgz<bp3&wE-ezFvQQ{PT`j
ziyr~+amN27@DCUNd-stx{;9k5X>_knM-~6Pm)F{V8oWmx|2)rW@uT2<g3Q!Rf4lN&
z*MI&Vp~b%g-gf67uj_x%1O6;{k2?OjB5C!%tJYs1|BMJN{v3F@UuHUv^?x3?HvWwe
zhkMk1f6?k+565F5k0QK|V@&dHmiLpq&+bI#JjwZ*`7FM9&qv>B_4^;s;ye$t&F2*A
JMf%!W;Qy?8Su_9u

literal 0
HcmV?d00001

diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index 4a981bd..a0626d1 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -63,6 +63,15 @@
     "AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation",
 ]
 
+MSG_TEST_STRINGS = [
+    "# Email Message",
+    "**From:** test.sender@example.com",
+    "**To:** test.recipient@example.com",
+    "**Subject:** Test Email Message",
+    "## Content",
+    "This is the body of the test email message",
+]
+
 DOCX_COMMENT_TEST_STRINGS = [
     "314b0a30-5b04-470b-b9f7-eed2c2bec74a",
     "49e168b7-d2ae-407f-a055-2167576f39a1",
@@ -232,6 +241,10 @@ def test_markitdown_local() -> None:
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_mskanji.csv"))
     validate_strings(result, CSV_CP932_TEST_STRINGS)
 
+    # Test MSG (Outlook email) processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_outlook_msg.msg"))
+    validate_strings(result, MSG_TEST_STRINGS)
+
 
 @pytest.mark.skipif(
     skip_exiftool,

From 08ed32869eae01d0b7c39944a092b90221f81ae6 Mon Sep 17 00:00:00 2001
From: yeungadrian <47532646+yeungadrian@users.noreply.github.com>
Date: Fri, 3 Jan 2025 21:58:17 +0000
Subject: [PATCH 3/8] Feature/ Add xls support (#169)

* add xlrd
* add xls converter with tests
---
 pyproject.toml                |   1 +
 src/markitdown/_markitdown.py |  27 ++++++++++++++++++++++++++-
 tests/test_files/test.xls     | Bin 0 -> 27648 bytes
 tests/test_markitdown.py      |  12 ++++++++++++
 4 files changed, 39 insertions(+), 1 deletion(-)
 create mode 100644 tests/test_files/test.xls

diff --git a/pyproject.toml b/pyproject.toml
index 67f6825..9c113ad 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -32,6 +32,7 @@ dependencies = [
   "python-pptx",
   "pandas",
   "openpyxl",
+  "xlrd",
   "pdfminer.six",
   "puremagic",
   "pydub",
diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index d209b5e..50c83b4 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -726,7 +726,31 @@ def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
         if extension.lower() != ".xlsx":
             return None
 
-        sheets = pd.read_excel(local_path, sheet_name=None)
+        sheets = pd.read_excel(local_path, sheet_name=None, engine="openpyxl")
+        md_content = ""
+        for s in sheets:
+            md_content += f"## {s}\n"
+            html_content = sheets[s].to_html(index=False)
+            md_content += self._convert(html_content).text_content.strip() + "\n\n"
+
+        return DocumentConverterResult(
+            title=None,
+            text_content=md_content.strip(),
+        )
+
+
+class XlsConverter(HtmlConverter):
+    """
+    Converts XLS files to Markdown, with each sheet presented as a separate Markdown table.
+    """
+
+    def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
+        # Bail if not a XLS
+        extension = kwargs.get("file_extension", "")
+        if extension.lower() != ".xls":
+            return None
+
+        sheets = pd.read_excel(local_path, sheet_name=None, engine="xlrd")
         md_content = ""
         for s in sheets:
             md_content += f"## {s}\n"
@@ -1353,6 +1377,7 @@ def __init__(
         self.register_page_converter(BingSerpConverter())
         self.register_page_converter(DocxConverter())
         self.register_page_converter(XlsxConverter())
+        self.register_page_converter(XlsConverter())
         self.register_page_converter(PptxConverter())
         self.register_page_converter(WavConverter())
         self.register_page_converter(Mp3Converter())
diff --git a/tests/test_files/test.xls b/tests/test_files/test.xls
new file mode 100644
index 0000000000000000000000000000000000000000..de4f368c24d489ff7aca786acc29d80652189eb9
GIT binary patch
literal 27648
zcmeHQ2V7J~x1U|wf+$5qipWY)L_m5`aVesxfY=K*$|6JrL{R)9fvCh9MKo5TF^C#t
z?_JbbP)t<B-ch5$9`hj@-FMF1UATL<0Pp+W`@OQ4-`u@(?wS8NbLPyMa_=q|e>1ti
zc8Api!a2GVE%I5SPjp#y9$b6N^2P-2OC*?;eR{z)kTl2tA`9pXpdod2$cTOy^o|og
z(G?K#06w>CV~&uAkVcYpVn|4ELPm0Is&KeCD|MXkzxoKFI>ho&hf;97K%O??0j&s<
z39+QQnbLDhdfrAk57ToXF(r~*L#?wT{C)kQJb{wK_(7EKTY9#m=N540k$tqxO-k_~
zZOLum7mxyUb`Y+|3nj56g(Q<Wk^yJrc>UkPaT|0Kki%#I6?!zJsWVA}TKLx0r8%cA
zLJ|Qr8b^{yCasqXj|bf9Bx*Ovfm*A_K<k;*>7>fOW({(5YmlRR0~|aeQd`^CP#5n8
zI4~lr_Z1$|ZxF+v0SrEA-2g{JzJgZQSBcS9yA9C+8N{yG&fdk|-hE8Rt|OgfnUNja
zYPW-tppD%9m;vI1QSwZ7M330iGr76T<3JvvI)Gc-ak)5XQdU3WOgsr0gQ_?SWtvH|
zBb_K=WXG;jf}N`!=sm_MXbebkBOT!nin7Z&3)zL7g>ofC#D#Q`7sk{;ULIAAB~m1+
z7wStoK~Jg|Dl1<nS5}yct}l6?m_hpx@|Q$NKbcdhP(w#c{QK>|BSOZ+c;++s81Sk6
zcS~<+ucG!>q;FF}pQD1lQU!gD3i>7$^vx>hs`Oc^g8yd~bXE9G<^Ol-vqOcR2PoZK
zp}iP>Mfx5U{QFeUcd4KsRY7O;ff*9Vf3{G5m|rpUB_<}znk{Qaw3&7>t)nL<9<x1W
zOSBn%bRmXNKYR}i^kDo`7s8oxx{C_Bvr;-n8jABy3NP(n&1CdC|7ZLVCWH+C6%tKI
zo`tMG>dVmyn4Qtnpq?&-OGP?IZVMTIoqZVpF_F=)kg;%SCBttnqu1%l_$RJZXuXO|
zOw5~^gPoavXC$KoK4WLb{<<(8Dzm4ajGv=tD;fO?ObE%2Iy<zK(K-A&K&NpA4TuNC
znP~zuGEJ7JL=mwIkE_iQxRM+>IxRq&!?>UYml%U^@PuK<xdn{We8mFD!f#3eW@6a{
z38+YuBxpCa0Gg~x60psiQ~(>dNd>S`np6P0MUx6(e`!(y?5IsDfRU+51+a@XsQ~uj
zCKbR=*`xy9n_8e#Qwz8>wSa3=3%E740L(_4W~Huj1+c@iqXsWW5X=cPFj<8FBMSoD
z0(^lq{$NjLCe}isA_&aT8wQbx)PvyWQG=RzddY)uqoy~2)Vq~Jbre;1uak=55CXe|
zZ7{wz)E{r0e)X$Fb^V>>K#GX2auCTGbs&E7AVXDXBB?H7X|jmZoIU`eN1GG17KtXS
z(4>4t9f+Q^in+PD5~1|KW=8-vO`k=BTl6qcLdy|wl?nh}wo)SHO1*viRzWF|NTi??
zn<i5VX;PhtOer=^Q7Kp&vXuIeqr%+Elww=uU$YdqH~NB7Odmk&D<~xsCduQ~SxT{H
z&UH0I|HnhAOP7?C)Pcx+fNfftFgC66=H_<HKv09}m{>DtD|sJKtc5?YF;0WEP?-P7
zI;KQYu274eJ9jEL4Vxyb1=0YU9x|t4(-h}eoq@N$7U&0Zrp*xy=JdyIP(ZsbyeiZJ
z?T*!B2Q{?2j3(=kigC*hO4Ub8$S5|l3`BL@!j(xqh=O)#V=W?1WNW9W2`;x91SyYD
z4+1+&4TErlKezYT%Yi^fxlM4AsoEyE>S(Y!;zaczu#wg<NO^>M5ZFj-7{t>{J&1!G
z2xOGo1n1GJZGtP+M%Y9>2&|VH1}TqF57M?VAf8_8K^)~kAfwzSxKvPW6I}H*!Y1lL
zEE@w-9-$t@wlN@{Ug|;G%Yi^fxlM32q}nECav&<>iFy!N=QnKF@(A@H!p4AjdZ`EL
zAO`{&<u<_upK6=nu0bRGL_LUgV?fFy)PuBZ42Y)}8)Pxxlo-SCMLgXdg|5Qj_;_(@
zrbnS5k`$upqz6yCt)w|@)4LN#A|$Smf=N95iAgGbu!IKS!2?7+ZE@nA>Kcg-UZZ!w
zEWi;rGf<2PSB$5k7&|p$m^BC$qa#lI`_Wb9VobSWycET>S0jd5rBE@t;=~s(UMLsS
zk}JkrQH--1F}RMA8ck1}D0y41T#Ol4jE|xiH#K5#xgZrI5GUSv@K(8)R$MW@iefy~
zh+(F+)M)zR#D}kLDi>ppV!YLRnk|U&ZiJ_mM;y8KQn{GcTro;K&0dWdwx^Xxyg$E1
zxflzs7$u(8L5&!;r<F(Cx^+vr7)!1gC7#wvjTpA4l}AXPtWYk-iYrEmr*&2%hV5zP
z5xY<Su3U^YSBw%*^HL**?P=u^r|<7pE~X8N@m23>LJ;HI2v76$T6N-yaxpesF-knm
zL5&!;r+Iq4+dfmdn6_LoN<7U;jTpA4d3u$Vl_?iv%N3)<(_GYuVSAdVm*hsGaxp@#
z7$u(Ou0{;o(>%Qv{G_7awBw3V;%VM$#IQZh)9a^;3zXHy5T0|xysK{94CaiRHWEFR
zV<jOmt11~#94O(Cj+KVQz8(}jiVNC~1IqRyLSE!}C<k?@2L=D(f;w_Q**-%E?j_Js
zovW#1Jt%k!7qlY>l<gr<<x-*?w0%7&_y!l$l>^H52SR#e{GuGRLp>;X0vFVS1IliF
zLejVNQU)p@PB6NXkl2JmL4bVA^Q<LWBm}k_6Y54_9o!Uw9YVAtUgI{Rw8f3#$scU`
zhsLJFXQjkuCa0wq3ci*H<eN*H5#~*$c#;M0E5*XpE19Ih6b$!KaO_1;+|H5-!0ja}
zK$pNHCAf+E89ZQS=LB*&eYxCXz7f2h0go?2lf?1kg`wh<ltMu+N3Yfr1JEl8CWG;0
zJWLHk;VK3G*jg!dADLFT|D;=|6_b-`#pGmKF}cOCY1=7JOHwEh%B!NI1yv6PlT3iW
zSWtiv<aNo@1zsl(uP(g$SkEgYF)$L0sw?A~rwwJ=bINE-1W+amUK7ipbKiI}wXTF)
zo*tBN<CM^o7%G&Im(_}Nh8Z{s%T7;;O^wY=%a|%0B+kw(6!emZ;`>O9Nf^n7H}$aX
zQ)&CBfg)4E=7T_;Z14x%yppv(Hn!Bg^r3FBJ<}^K4IH(rJTA{lqD^{%(b71siaX*m
zED$EcY5}jmzzQN<9Gj2~ZwbO$fupFFF<3Sn6hhxm21Un#g3MD0gwul#(__#V3>><m
z9%#k{!o~a+u*hJ6L+s;%n<xQi;-3m|iD+~AA5akl_hN%X4CI2FDgkHWpbBtiYQU%9
z-fVD)g`9w()lvyK#zUsX<jk^|Z$NyYAxZDl^ek|NEcqtmn?ZeggJsjf7qQXnyT1hX
z{bh}Y4+s9m?r#heB#289$*JST3DW-Gm!pgY#3g*oFco^g7@C|N0krD8u`U+Lq7QCy
z>w@77i67`sqQqI58L<#vIDwhh1_E;w0Ao!vz^5><Fe@4wPvK#StiDC?L`CZG;8J-6
zx}%r8t~>%FqrlTMXn;T{sr7kq3DLQ(zxZQ+iAu{97YgF!G57}1Ul1b`lp;_+S>(ly
zYFS9bP3HPQ%j9GsjmgPE8k1YhZ$<(LY-;t(%A_qmg(I{TK=-3IP;7H{$j1mN*XBs5
z*M!N*G+}ZwO_&@8%r1F4N`jvb1ixsga|X1qZ0V0iLuc)Z&N?VLGX?DBA+NfY5j4;s
z@Q+MrI<%Ogzw;2H8-+;AL?7VnK4@7|2sAm<{@T>7<aQX)^0G+6)K`{6Co>^|5PHba
z*o;&RT1<cy$t%il2C;f5NXP(Jmc=b)ffa{wYU2fhVPpYoo1_40X99nN03?pqb|9!c
zm0m?c-7y?yf#-RG9h1pq>O=J$>^F6N$yoZ%az51R6h0#c-x!!A7H4|olMbXJ7%8^I
zJ;2@9-NVz{HO@!u=j!d};pZBV;O^(@74PotD~?+!upld0>M=gfJZdS1m%@u#_YiPC
z_&g8ZkUY*4KqAEFj~3)_vIxh2nxIF7rlkZkA0f;~DD%++nm;%tJt-ElA>zzf$o7hz
zFabY%h*L6S9pJ%hVq&~sLTrMok7q)>t9M+0ziU984`e;WzCH>5vHr0M?!e}onCR`{
z=N0Gb?HTI>b@KNHHje<;MDKujfbdQT@Jf)t+us7x8y>bpo06IE*$Y171?G_8LVB1>
zKlx;-U~#jhtUGuky04bD8CVs`?Z~scM#rF8NgH@})e=5qLeo++VJB?NpsDHNNu%7d
zC!}m%bUx~&yKz`e_}?=gyY?yC86<FgHvP<kGe1r%J-foeyzKhU4TC?{L|h&evQcQB
z;&Q{QW_8!9j3c&sdrmGd+R$g)+mKEU)z-a6mE7~|cVo+tIbT~@bc-!9TK?|f(Zf!D
zv+{b+SvjEO%O~mMB6bEYD9EraSa<n@KmWYX<o0==@>*QT>VMm!dd?#Ms<O~)kG1x0
zw;2}r@^S5Qhld5b4~l%FHunh?%-B_0c;)%V+MZ?RA#Udmbok3{U8gx~FAP61sMd1(
z)6rs|OFkR#3g0%HZ+ddO;hZtQjT2hBJ~`HP^_=Iot`1K*d}qPV1=D;-mq#6!n2r5Q
z^t5F~d2UQDSVRI-S$TVvT;1?{DcH#u+yq}%8IvJSnbg@0&odYP_Qgqe!&$HLkIcA~
z)Bo~G3&+WwOA10e7w3=%PPs{T*0-#yZ}mH)Yj8*~bKR?&p8`K`omcx{$?rn*m78==
zJ`5F|PkJyp@L2Bjvix-$90f7i>E9H8d27#@!_k+DyIt_GE(zc3{++4O>>K?K57-)#
zaH-7ra-XnU-f<4O8x~C*bIo@7YRmh+5fy)#54h$xcI(28Ws+Sd-OE?^JD$7E<#CvH
zs^5}`<VSI9H+HTXc%)#Xc+-`(7x(P1z4{5NqFB-KQ<s|0fC|oqgpaT?lf)CmomuBd
zn6!-(XP}1|mPZ<#6dDIT$XPq3;D%rPA*Zmoo#(e6`X>o3vI;))!{s%%u05`r;rHu3
zv&vO{n-6y>(rZ3fe8Fw@zSq-SHdt@G8Q$iUOUi>Yt)4u~wkX^EZOePT&ra%}V_#X2
zzh%!I;nh8zD%-tsxik8J!xw?`dJTPAQPc5ZSaqlN27G_z$`@C==Qe{uWimX4z}~=8
zo2h~as{MTUg?ZL~X2)885E)KNOL_Ide)PlUhd!U(dm}RB`MO6=zn%4(>r)(DIHfA4
zchnIxgXCu=7h{Y{ye|fYcK#G^ZhG_G!l#Q9yDd!#_;`H4<w75qwijLnetLIz)xE{1
z{ZhO4-Z*9c0inx__j|?<zIQUYG^DVnjpTUJ>$0qSElo?>{pfPaCU-%8?xdQkg{eQ-
zSXOC&Id;uln=5bPJMQ|vPnfp*{9k;QJ}`S3Y0<A>X5Y4hLl#7priFBhsr+JBp|J47
z=((opUrzCC$NOi`b0x<PrQh9<zM{`sUH-XQ=Y00Mc>5MzH47;ZTbA+4;?e@6KW9Eq
zn7JnWUU7K%{!4Fu{_xwj=XW>$@coI|=bG>R=1!2L)#UtO|2vl_y-gY!HDa6J9~X{X
zKBgOeU_tf14;Q}L^=<>tNien~pmoK#_+v-zjW-*nwIny|{mVW7xaHF<?ft#cRfn%d
z+cw)4d_eNS>MQrBA!e?JbsxV!ZFO{X+JvlX^U*UhrcG*oZ_~@a{%{2@565Sj%|}h}
z-gjtT<<bKEV!f#BL;Y=c?=|?nui2o`;iJBotKawW`VoCE#-G+B(Lr+^dItvl_F7vf
z`qZc%Wy@Q>D(K;1Hu{?Z?Z@svIM`;;z6ZxXT=~n-#$@-5MX_75c`bMNu6v(tci&;L
zjl}EqAJzrewvGP2d(PAk@Ag%U`z>nC=K<wgMh%~==k#sFp?#vSKGoEW$<F=s;pDB~
z?>!p0?P30$+&}+ZmiCpTYShcvQ|F@(fBxsGRm(b$&aJ6jBMO|7eXlAh<wP%s<0lTy
z=y&*|%hx(x-dA}XTYk!^=Px^hJl;><|ML8G{@oS!f1G#`-*u)@z{dH_vd91P`ok{s
zYa>k0l<%whh2PerXN~0)-TQl2o}c|XX=%@wWwqtM-0Ai7y3dr>B}1lv+-3OayZs|4
z$ERF6+^fHH|Cs|_0<#uIdQ9H@(xlyxkqg61SB!1>EYR+5ce`H;=AXUpysqRDueY%C
zq}IXVds?k3o^`i-@z%jLuXlVq^9;YgUaiUcAr@(CSM->3{;|OEisAcqg*)awO8-@>
z<+awPo;I=P4u3T}KmEq-$J<BSKid}^dplY5<Db9u%;>$|ZOyFN9xZRgSESA_EYG(o
z^!DAcW^TtNPH$g&JezLo@ln73wF12*sS!1y)3;=<an&{o8X4K=Jiq^x#M_I`$1ETI
zSJ$En{Sza^K_6Xv_Y42W`lH*NzO1^}fAX1wTYFaAS+HgE$mk0-{u8?o)azt*RKLb{
z(95<fOKRM^xm15&JtIBWI8~C{`Sz;SkAGfPT@o@Q`{%&%M~7bPT9Z(9Iw!=*=3Kz3
zSs}0P=)5aQCo|4P*^Zbvwcvc*`%SA8mY1FgYm;of(5p6J(Hr-kmMO2tAI&hGb0W*(
z@O=I2HPx{j((*U#dNez~cIuuX1IJa2?516K{>>M~ZD#HIru@={t8Wd59K4odpEG2%
zarEBS?P4Pm7hj)TkmRa6xZqt`o9R!4YllvDbh+8W{MvUP!!}(VM)r7i+CK5^*h7_X
zth3yC_WlFf%?<r-L$}I#&;LjmEne@ZQ?ZF$4|8!IyW-j2hX*H4{UU4XPv6^DnClnc
zPBHD`d2rSoi~Ri?qkR6+)~fY|tS?s#n77&>f3;Dc@x6<)K8<p5TQz=rOr`G1m3@K-
zzFJr1x6eA-lXtGmXYF4<z4j^lVe-tIukEakpWhYq_V2Hc&EJ_Y@@!kviEYzeyBx4`
z=n?m_e_`gdJ%4*P_pY@1@j&I?Amh6+850V=tNFH#!<{J4)e-5!3%jlDoiEoqI!)W!
z>eoXz|9(4GT$WhyGPcd??D#HKH6}}j6%QWBGn|umO|+-m%64DZtSBjPtxhZPXti_v
z(da8aOBUL%Jhkzu)B5GVr@KCOw8<HKx^%zuRYRUn$N=vX6W3X7wjA{Hs->gsf8V~N
zdi8{LcH?G0**R{+rXLTkk2$s3(tN^L$EC~0zt#OYY<6%(d5blDYc@UF+j_*lo&B5+
zX=@*zzoxi)b#Xh3D@zv-zhYLEWjSQ!sQEXK#P2Be+W7s3-WHqgjod$KPRtGaqKf19
zOS*`zZyDlh7-{~ok4^HYle21XX<LhQ-#k4YbKmm2kPW6^O}#Jh?GbvULq~Dws+XT(
zO|CfhHNRA_W-kmI)-W>SlT!BZ!3_A6!Wsdk6M#sA6Ygd+Yo<>Tq@`6HG+xyH&g_d@
zu4v7C<kQ?&<k0{5i{@Kh1a7lpkFH2QwCww`mhql<J(4zA3A#M{ZsN-06ZSt}5H!F%
zYR-+W3ERg`Uy$baqSd%XeVroPI}C7g^H`qIqC=m>MHlA|s?FQ#U%YV0BfH}D!DkO$
z46n(``Qgnf-n3H=N8*gXtR$98qgP6Hee$=OGp@6(`KZtRgD2g{*KzvsboqVzRhuHT
z&&3=(_{%-07*ll~-xXemQ8gn3HrAnJei+?X9)LSVsxR#h_GC?%xCOaWj{1U@Fc)Jt
z0@wUpB<}cMk91WT5}gfjF}O}pkI@P?YS`MSaf#@@I{3l{tlD(QD5$+LE-A<=Sc9x$
zP0Qg1nGdx2EE@75GxH@@W4SM}HshP&ZF=ClC%53;*nnUH+y3~aPeV8-5F>hR0%wp4
zll{McdUH84E?P8((w!;26FpC-lsvj1U@#qFDNKk9?`ybo92N_Nw1PB)a?hma_Q0J;
zk0&Dh&YSGosSf3-s3YroEj%sI0z@9@$HN^4IK1dlA(DL>9V>!;$0CjmD690;4Y#D|
zF&MNkAUSjyM0k|~eCHYt*y+LpZNj{HCg9_XV0i6<MusJz5kVSexHPmBmj>ZMfri?1
z`Cz%HK*NUT^1*(O0u3w7<pU2>prJQ#`EUusrQy<$OM_>E3N*A8mk;g8rJ)76G<=ne
zOEXnM!zCq`5B-TtL;vE^aN^FT;oE0y+9%pa2<Z+~LYN#6a@ZbBjt@C(2_~lnIc$R?
z@KtB@0jO`#kRUkc4Z<@SPR~w3T9gAl0;vO$D&-5J9AQ!p^bh2SqSV3k>?J)j7^oL=
zv_Wc6e|pb>o;{=(s4LtnrKP1(7^J>Z4%8np{Af92JTrVGK+1u20%{(E&y?<me0h;l
z4)ij}m(qIXh0uFpcqTql4)i(XU@+Y1J!%^gB;`O4#1h{09>r&_OQWS6=#PjIO*uU2
znL!Gda-er21|!?7o`Yh11^Z#>tH==<gl*Ow+Q|qvn4v}R4HrJW=S$1kNiorrp+sId
zQiF!ldlvK@D&;^t<X~F48@=~hgcKPEdO74vY3aPal!NLIWtf&kpGTbD^j<GK=MAIh
z!BP(F0mva@aWH%`4D1ic(E~Yx7%MUUB9dZY?|^%Z9#W1#iq}udfqexrXglRa(0jr3
zEb|xaImp3iN2TO3t>q%cz<z{M7@wncW+W2S^cuW};tOtB&hv-3t^ChZ3B5btl<cMB
zT@c{ofm+Oik!_w_R5yTcQX3I11KLe_(8zG5vjUp7QZP=?nv7#H!+?g*hqZb%ZZyQc
z)GQQc8NmNPcQlNI=%XdX_=70{(l7`$m}E2@FL`i>887*8Zx~-TPRc%cFjoG1#!Ivs
z##Q)ni2sK1l8M9^v+-bZ7_+hNOb%l<#$G0eF&pD7D`pegccBNNe+lT>SnAnW4yjCJ
zVWI$yy+kPGki~2}T8@du?ddfW_bsR;5W|^LUFg{@h)?y=23=qd0e7##7B$W)qKT}R
zQ2&QJpcp+j;QywyT097AaHZ8K|ChpQJ{n_z&#<v=#7DQ1Kag@if6d$4eo4kZFA|VW
z9d_Mt2ijpsIFRD~AV}CD!XaT}kA;NcJsT1ZVMUN|pxOutgZger=z7K-2y||oa^sK6
z#7+?QOv}nh71O8X_4N85n*lb;k6Wyf$?+L!lhP71g^{uG$*GxXlahpC+415OxEBP8
zWKX(%Xw5|)UguptzW4TkTRe1T&7rY?#sV4(Xe^+yfW`tE3ur8$v4F+`8VhJFps~RJ
zXA7uX|MM#^R2I2e8!ui4>wlNpt+@Vw3JEuwas3|+3AccTLBh>a+@!<JZCs1v+CCE!
zuKTA!!tK_XkZ?a>J|x^9I0=cpAAoxacrfdAV~T}&#KeR9*%pLgdEAd+-hU)QxNP9X
zM`9=~hFc8?IW#R}{3LvRauSt{h?+xV0gVMT7SLEgV*!l?G#1cUKw|-o1vD1WSU_We
z|63Nobv&+kam|hEZ+y)K-%i3cH9phFwLZSmh6`m}v*Y?*`YIV+<Kqio_{<;I`S@BE
zzIcUeecTVgS4`o{MflU*1b@N`pWoxJ+Tp_67LpKBJ4kkr@P7z7z^5am_K@He0r>d@
zg8#w+eq?~T?*gB$klY}3hU5+jzdG&-3Cn+igrAzD@fH{W-!sLp&I;kz1Mug%@D~G`
za+nhx`tlh{L_ZP=nEjJd2BfDlEZP5y!h1tOdIo$qJdWxyUKzVu8vGBSI{B!BT4vVG
ziXKcxhYY0Oy2YP_z`ElX=M$+;NU8q~4OUUPODB+zeF*zcW98c;0&bAOp+kCrH2lqu
W3Gf|YY=Nff<Nu$ie<u_ymH!`&m{tJ*

literal 0
HcmV?d00001

diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index a0626d1..1ac9041 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -54,6 +54,12 @@
     "affc7dad-52dc-4b98-9b5d-51e65d8a8ad0",
 ]
 
+XLS_TEST_STRINGS = [
+    "## 09060124-b5e7-4717-9d07-3c046eb",
+    "6ff4173b-42a5-4784-9b19-f49caff4d93d",
+    "affc7dad-52dc-4b98-9b5d-51e65d8a8ad0",
+]
+
 DOCX_TEST_STRINGS = [
     "314b0a30-5b04-470b-b9f7-eed2c2bec74a",
     "49e168b7-d2ae-407f-a055-2167576f39a1",
@@ -185,6 +191,12 @@ def test_markitdown_local() -> None:
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.xlsx"))
     validate_strings(result, XLSX_TEST_STRINGS)
 
+    # Test XLS processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.xls"))
+    for test_string in XLS_TEST_STRINGS:
+        text_content = result.text_content.replace("\\", "")
+        assert test_string in text_content
+
     # Test DOCX processing
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.docx"))
     validate_strings(result, DOCX_TEST_STRINGS)

From 731b39e7f5d36469b2912ed1608fd86c04a1ddcc Mon Sep 17 00:00:00 2001
From: afourney <adamfo@microsoft.com>
Date: Fri, 3 Jan 2025 14:34:33 -0800
Subject: [PATCH 4/8] Added a test for leading spaces. (#258)

---
 tests/test_markitdown.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index 1ac9041..9dc7374 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -257,6 +257,11 @@ def test_markitdown_local() -> None:
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_outlook_msg.msg"))
     validate_strings(result, MSG_TEST_STRINGS)
 
+    # Test input with leading blank characters
+    input_data = b"   \n\n\n<html><body><h1>Test</h1></body></html>"
+    result = markitdown.convert_stream(io.BytesIO(input_data), file_extension=".html")
+    assert "# Test" in result.text_content
+
 
 @pytest.mark.skipif(
     skip_exiftool,

From 436407288f01b5a2c31111062b0c2ac959dad443 Mon Sep 17 00:00:00 2001
From: afourney <adamfo@microsoft.com>
Date: Fri, 3 Jan 2025 16:03:11 -0800
Subject: [PATCH 5/8] If puremagic has no guesses, try again after ltrim.
 (#260)

---
 src/markitdown/_markitdown.py | 19 +++++++++++++++++++
 tests/test_markitdown.py      |  2 +-
 2 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index 50c83b4..aceaa86 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -1594,6 +1594,25 @@ def _guess_ext_magic(self, path):
         # Use puremagic to guess
         try:
             guesses = puremagic.magic_file(path)
+
+            # Fix for: https://github.com/microsoft/markitdown/issues/222
+            # If there are no guesses, then try again after trimming leading ASCII whitespaces.
+            # ASCII whitespace characters are those byte values in the sequence b' \t\n\r\x0b\f'
+            # (space, tab, newline, carriage return, vertical tab, form feed).
+            if len(guesses) == 0:
+                with open(path, "rb") as file:
+                    while True:
+                        char = file.read(1)
+                        if not char:  # End of file
+                            break
+                        if not char.isspace():
+                            file.seek(file.tell() - 1)
+                            break
+                    try:
+                        guesses = puremagic.magic_stream(file)
+                    except puremagic.main.PureError:
+                        pass
+
             extensions = list()
             for g in guesses:
                 ext = g.extension.strip()
diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index 9dc7374..e2d2e75 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -259,7 +259,7 @@ def test_markitdown_local() -> None:
 
     # Test input with leading blank characters
     input_data = b"   \n\n\n<html><body><h1>Test</h1></body></html>"
-    result = markitdown.convert_stream(io.BytesIO(input_data), file_extension=".html")
+    result = markitdown.convert_stream(io.BytesIO(input_data))
     assert "# Test" in result.text_content
 
 

From 05b78e7ce18cf2f8d8d75058a1f2c98f9930318b Mon Sep 17 00:00:00 2001
From: afourney <adamfo@microsoft.com>
Date: Fri, 3 Jan 2025 16:40:43 -0800
Subject: [PATCH 6/8] Recognize json as plain text (if no other handlers are
 present). (#261)

* Recognize json as plain text (if no other handlers are present).
---
 src/markitdown/_markitdown.py |  5 ++++-
 tests/test_files/test.json    | 10 ++++++++++
 tests/test_markitdown.py      |  9 +++++++++
 3 files changed, 23 insertions(+), 1 deletion(-)
 create mode 100644 tests/test_files/test.json

diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index aceaa86..b6acfe8 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -173,7 +173,10 @@ def convert(
         # Only accept text files
         if content_type is None:
             return None
-        elif "text/" not in content_type.lower():
+        elif all(
+            not content_type.lower().startswith(type_prefix)
+            for type_prefix in ["text/", "application/json"]
+        ):
             return None
 
         text_content = str(from_path(local_path).best())
diff --git a/tests/test_files/test.json b/tests/test_files/test.json
new file mode 100644
index 0000000..eba3059
--- /dev/null
+++ b/tests/test_files/test.json
@@ -0,0 +1,10 @@
+{
+    "key1": "string_value",
+    "key2": 1234,
+    "key3": [
+        "list_value1",
+        "list_value2"
+    ],
+    "5b64c88c-b3c3-4510-bcb8-da0b200602d8": "uuid_key",
+    "uuid_value": "9700dc99-6685-40b4-9a3a-5e406dcb37f3"
+}
diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index e2d2e75..3333bcb 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -145,6 +145,11 @@
     "5bda1dd6",
 ]
 
+JSON_TEST_STRINGS = [
+    "5b64c88c-b3c3-4510-bcb8-da0b200602d8",
+    "9700dc99-6685-40b4-9a3a-5e406dcb37f3",
+]
+
 
 # --- Helper Functions ---
 def validate_strings(result, expected_strings, exclude_strings=None):
@@ -257,6 +262,10 @@ def test_markitdown_local() -> None:
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test_outlook_msg.msg"))
     validate_strings(result, MSG_TEST_STRINGS)
 
+    # Test JSON processing
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.json"))
+    validate_strings(result, JSON_TEST_STRINGS)
+
     # Test input with leading blank characters
     input_data = b"   \n\n\n<html><body><h1>Test</h1></body></html>"
     result = markitdown.convert_stream(io.BytesIO(input_data))

From 265aea2edf31bf1b022992e59f0ade1e54903aee Mon Sep 17 00:00:00 2001
From: afourney <adamfo@microsoft.com>
Date: Mon, 6 Jan 2025 09:06:21 -0800
Subject: [PATCH 7/8] Removed the holiday away message from README.md (#266)

---
 README.md | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/README.md b/README.md
index d2314c3..6bc91e6 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,3 @@
-> [!IMPORTANT]
-> (12/19/24) Hello! MarkItDown team members will be resting and recharging with family and friends over the holiday period. Activity/responses on the project may be delayed during the period of Dec 21-Jan 06. We will be excited to engage with you in the new year!
-
 # MarkItDown
 
 [![PyPI](https://img.shields.io/pypi/v/markitdown.svg)](https://pypi.org/project/markitdown/)

From f58a864951da6c720d3e10987371133c67db296a Mon Sep 17 00:00:00 2001
From: afourney <adamfo@microsoft.com>
Date: Mon, 6 Jan 2025 12:43:47 -0800
Subject: [PATCH 8/8] Set exiftool path explicitly. (#267)

---
 src/markitdown/_markitdown.py | 39 ++++++++++++++++++++++++++---------
 tests/test_markitdown.py      | 32 ++++++++++++++++++++++------
 2 files changed, 55 insertions(+), 16 deletions(-)

diff --git a/src/markitdown/_markitdown.py b/src/markitdown/_markitdown.py
index b6acfe8..33806e1 100644
--- a/src/markitdown/_markitdown.py
+++ b/src/markitdown/_markitdown.py
@@ -892,14 +892,25 @@ class MediaConverter(DocumentConverter):
     Abstract class for multi-modal media (e.g., images and audio)
     """
 
-    def _get_metadata(self, local_path):
-        exiftool = shutil.which("exiftool")
-        if not exiftool:
+    def _get_metadata(self, local_path, exiftool_path=None):
+        if not exiftool_path:
+            which_exiftool = shutil.which("exiftool")
+            if which_exiftool:
+                warn(
+                    f"""Implicit discovery of 'exiftool' is disabled. If you would like to continue to use exiftool in MarkItDown, please set the exiftool_path parameter in the MarkItDown consructor. E.g., 
+
+    md = MarkItDown(exiftool_path="{which_exiftool}")
+
+This warning will be removed in future releases.
+""",
+                    DeprecationWarning,
+                )
+
             return None
         else:
             try:
                 result = subprocess.run(
-                    [exiftool, "-json", local_path], capture_output=True, text=True
+                    [exiftool_path, "-json", local_path], capture_output=True, text=True
                 ).stdout
                 return json.loads(result)[0]
             except Exception:
@@ -920,7 +931,7 @@ def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
         md_content = ""
 
         # Add metadata
-        metadata = self._get_metadata(local_path)
+        metadata = self._get_metadata(local_path, kwargs.get("exiftool_path"))
         if metadata:
             for f in [
                 "Title",
@@ -975,7 +986,7 @@ def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
         md_content = ""
 
         # Add metadata
-        metadata = self._get_metadata(local_path)
+        metadata = self._get_metadata(local_path, kwargs.get("exiftool_path"))
         if metadata:
             for f in [
                 "Title",
@@ -1036,7 +1047,7 @@ def convert(self, local_path, **kwargs) -> Union[None, DocumentConverterResult]:
         md_content = ""
 
         # Add metadata
-        metadata = self._get_metadata(local_path)
+        metadata = self._get_metadata(local_path, kwargs.get("exiftool_path"))
         if metadata:
             for f in [
                 "ImageSize",
@@ -1325,6 +1336,7 @@ def __init__(
         llm_client: Optional[Any] = None,
         llm_model: Optional[str] = None,
         style_map: Optional[str] = None,
+        exiftool_path: Optional[str] = None,
         # Deprecated
         mlm_client: Optional[Any] = None,
         mlm_model: Optional[str] = None,
@@ -1334,6 +1346,9 @@ def __init__(
         else:
             self._requests_session = requests_session
 
+        if exiftool_path is None:
+            exiftool_path = os.environ.get("EXIFTOOL_PATH")
+
         # Handle deprecation notices
         #############################
         if mlm_client is not None:
@@ -1366,6 +1381,7 @@ def __init__(
         self._llm_client = llm_client
         self._llm_model = llm_model
         self._style_map = style_map
+        self._exiftool_path = exiftool_path
 
         self._page_converters: List[DocumentConverter] = []
 
@@ -1549,12 +1565,15 @@ def _convert(
                 if "llm_model" not in _kwargs and self._llm_model is not None:
                     _kwargs["llm_model"] = self._llm_model
 
-                # Add the list of converters for nested processing
-                _kwargs["_parent_converters"] = self._page_converters
-
                 if "style_map" not in _kwargs and self._style_map is not None:
                     _kwargs["style_map"] = self._style_map
 
+                if "exiftool_path" not in _kwargs and self._exiftool_path is not None:
+                    _kwargs["exiftool_path"] = self._exiftool_path
+
+                # Add the list of converters for nested processing
+                _kwargs["_parent_converters"] = self._page_converters
+
                 # If we hit an error log it and keep trying
                 try:
                     res = converter.convert(local_path, **_kwargs)
diff --git a/tests/test_markitdown.py b/tests/test_markitdown.py
index 3333bcb..689d6f3 100644
--- a/tests/test_markitdown.py
+++ b/tests/test_markitdown.py
@@ -277,9 +277,29 @@ def test_markitdown_local() -> None:
     reason="do not run if exiftool is not installed",
 )
 def test_markitdown_exiftool() -> None:
-    markitdown = MarkItDown()
+    # Test the automatic discovery of exiftool throws a warning
+    # and is disabled
+    try:
+        with catch_warnings(record=True) as w:
+            markitdown = MarkItDown()
+            result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.jpg"))
+            assert len(w) == 1
+            assert w[0].category is DeprecationWarning
+            assert result.text_content.strip() == ""
+    finally:
+        resetwarnings()
 
-    # Test JPG metadata processing
+    # Test explicitly setting the location of exiftool
+    which_exiftool = shutil.which("exiftool")
+    markitdown = MarkItDown(exiftool_path=which_exiftool)
+    result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.jpg"))
+    for key in JPG_TEST_EXIFTOOL:
+        target = f"{key}: {JPG_TEST_EXIFTOOL[key]}"
+        assert target in result.text_content
+
+    # Test setting the exiftool path through an environment variable
+    os.environ["EXIFTOOL_PATH"] = which_exiftool
+    markitdown = MarkItDown()
     result = markitdown.convert(os.path.join(TEST_FILES_DIR, "test.jpg"))
     for key in JPG_TEST_EXIFTOOL:
         target = f"{key}: {JPG_TEST_EXIFTOOL[key]}"
@@ -341,8 +361,8 @@ def test_markitdown_llm() -> None:
 
 if __name__ == "__main__":
     """Runs this file's tests from the command line."""
-    test_markitdown_remote()
-    test_markitdown_local()
+    # test_markitdown_remote()
+    # test_markitdown_local()
     test_markitdown_exiftool()
-    test_markitdown_deprecation()
-    test_markitdown_llm()
+    # test_markitdown_deprecation()
+    # test_markitdown_llm()