From add39a921af765cc835f778ef2f33ac69c91acd6 Mon Sep 17 00:00:00 2001
From: Peter Kraus <peter@tondon.de>
Date: Tue, 16 Aug 2022 20:04:36 +0800
Subject: [PATCH] `extract`: Fix extracting from sparse datagrams if key is
 missing. (#51)

* Comments in electrochem

* Allow newer dgbowl-schema

* Sparse extract test

* Return an empty dict.
---
 setup.py                                 |   2 +-
 src/dgpost/transform/electrochemistry.py |   6 ++++--
 src/dgpost/utils/extract.py              |   2 +-
 tests/test_extract.py                    |  11 +++++++++++
 tests/test_extract/ref.sparse.xin.pkl    | Bin 0 -> 3520 bytes
 5 files changed, 17 insertions(+), 4 deletions(-)
 create mode 100644 tests/test_extract/ref.sparse.xin.pkl

diff --git a/setup.py b/setup.py
index 3859340..c3357b8 100644
--- a/setup.py
+++ b/setup.py
@@ -42,7 +42,7 @@
         "chemicals>=1.0.0",
         "rdkit-pypi>=2022",
         "yadg>=4.1",
-        "dgbowl-schemas==106",
+        "dgbowl-schemas>=106",
         "matplotlib>=3.5.0",
     ],
     extras_require={
diff --git a/src/dgpost/transform/electrochemistry.py b/src/dgpost/transform/electrochemistry.py
index e6db26f..f85d632 100644
--- a/src/dgpost/transform/electrochemistry.py
+++ b/src/dgpost/transform/electrochemistry.py
@@ -213,6 +213,7 @@ def charge(
     ----------
     time
         An array of timestamps at which the instantaneous current was measured.
+        Defaults to the :class:`pd.Index` of the :class:`pd.DataFrame`.
 
     I
         Values of the instantaneous current.
@@ -273,9 +274,10 @@ def average_current(
     ----------
     time
         An array of timestamps at which the instantaneous current was measured.
+        Defaults to the :class:`pd.Index` of the :class:`pd.DataFrame`.
 
-    I
-        Values of the instantaneous current.
+    Q
+        Values of the overall charge.
 
     t0
         An optional timestamp representing the time at which charge was zero. If not
diff --git a/src/dgpost/utils/extract.py b/src/dgpost/utils/extract.py
index 5d07bb1..8b64631 100644
--- a/src/dgpost/utils/extract.py
+++ b/src/dgpost/utils/extract.py
@@ -142,7 +142,7 @@ def _get_key_recurse(data, keylist):
                     for i in ret
                 ]
             ):
-                return _get_key_recurse([i[key] for i in data], ["*"])
+                return _get_key_recurse([i.get(key, {}) for i in data], ["*"])
             else:
                 return [None], [ret]
     else:
diff --git a/tests/test_extract.py b/tests/test_extract.py
index 7d6d4e1..8fcfb8f 100644
--- a/tests/test_extract.py
+++ b/tests/test_extract.py
@@ -141,6 +141,17 @@ def test_valid_datagram(datadir):
             },
             "ref.ca.single.pkl",
         ),
+        (  # ts12 - sparse extract with arrow and implicit
+            "sparse.dg.json",
+            {
+                "at": {"step": "a"},
+                "columns": [
+                    {"key": "derived->xin->*", "as": "a"},
+                    {"key": "derived->xin", "as": "b"},
+                ],
+            },
+            "ref.sparse.xin.pkl",
+        ),
     ],
 )
 def test_extract_single(inpath, spec, outpath, datadir):
diff --git a/tests/test_extract/ref.sparse.xin.pkl b/tests/test_extract/ref.sparse.xin.pkl
new file mode 100644
index 0000000000000000000000000000000000000000..ce9114b248149a2a43c2cb42ba3563e8ff56aad6
GIT binary patch
literal 3520
zcmb`K&5ImG7{+_Pb|;&-VI#(sNc=!?mxRGY(LJa`G9hd1q$LJ|1VdBZU9;7lp6;Q#
zdgDNXo5SoTP~c%pQ4e|$1VN8tPB{qP;>kn)f!yLj1VK;nt?r(go^)7hXY6JBqvoyO
z`_@z4#Zw!#&whGBOAfx!MS+8&h8<GeXj9a|g6}zn7&@KYZ;Ry{;<{MkRkO231wWh*
zcLRoLfc&V@K>=FBG!lH@Nk6o&oJ*g@HuOB0buoR{iW<W{g6}uupzGRKd~8~>6GZc;
zg1FP|rPXv|pSg%q)B~(Zu^k2xqp{6oU@Zw0d={!tLbipj^+l^>i7oMgSkg=ISKSn3
zKd(8g*Tu3^`TZ4f>t_0Ug_o`g!!X`^2uE_#n@m`@E<?!&;=sm~At=_xsWyVwFCyw9
z%a^1~lzB4<J1!_-t|LFjVzv46+}qu6@6AEIz8hdf%`T!$kmvZq+2pwub}Z<?+%Onm
z$cwAZKhxQW`oiKBhSr3^>t@6p)4}Vq_k0WazRZ1rIj5j$Tdbx%G#K*@5nJSiv?TJ<
zum<unc_nEi$Ho$nBL+E28an{K|56@bjQ}gt0-6c{4C?c`6}vF&0lZ1I$aifiwyw#^
zl(VLn<(cdy^JGEJHhFS<t`EuC%S-X~?eh=%hi{KYazUYx9RZhi2`FfZrUeiMfYg$l
z0olrv2{(ATM#A+JvgU*<0yY!?tDpRncHEJ)r9jFADC?EU1Ze)T^!G2HezZIi;0uMU
zxw4CZZ*~i~E05ns6?``>;70{OCR|0YPAFWq0*=bQqmVT(TmkUwE&+v4?(VdJ-xUCv
za8<oFp>Tz@>b^qOyl@4;gIxj&!u>TZ;9mtmCR|OQnM}CCQsd2)G)*CKPP-ywL5-ob
zVtI>duuO>=eReV>3TuwHtd=;ZM3M381dK<PF7M6%f$@$SBNK5}ujfT%Li)PJarf%r
zkHOtpC<4zrX9oDHc_x51upIoE$3-TF6hhBQV>$?x=6J@(>)4NE4$puxy|ZxQwR}j`
z0Z%Q8#-$7{Cb2ENgailoD(mOfKpJr)-UPdo5-$P~Qh6W3wMeNBDdiL3(yOi4-dzE^
zYaExJ_jtne5cEumnAhyk!AZURK+bW)EvBw5_+GP3!KudWHl`rQQmZb%kP;k%Eb_8@
zY-FR!v*RXOgZlM)AI8!WOqL%*UI0`|T3P-JuLlh>!n+1HG6eHIQAk-x9mr;YwTHnc
z#Tf=3Nx%<1Zcc_J?>p^>h`sT8PV_{qH2>$%4?b~C+vxZ1oq)%Ko9Xq3?0Rwwo|Eyl
z$F>P?J+e(mMhJ=z>(5IoXACw=u;YMzl$`Uyp33ZChLV@yGI(3W80?m=!(AM>Ox{I&
i#%%j&Eoprk?0$GPGM%^!CKxeng8P?m*v%-m8vg;JY?-+L

literal 0
HcmV?d00001