From c8e049fc680e6de158267c963296f27910f76213 Mon Sep 17 00:00:00 2001
From: jinye_huang <jinye_huang@foxmail.com>
Date: Mon, 12 May 2025 15:44:54 +0800
Subject: [PATCH] =?UTF-8?q?=E5=A2=9E=E5=8A=A0=E4=BA=86tags=E7=9A=84?=
 =?UTF-8?q?=E9=87=8D=E5=A4=8D=E9=81=BF=E5=85=8D?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 scripts/extract_and_render.py                 |   7 ++++++-
 .../resource_loader.cpython-312.pyc           | Bin 8341 -> 8642 bytes
 .../tweet_generator.cpython-312.pyc           | Bin 38150 -> 38297 bytes
 utils/output_handler.py                       |  14 ++++++++++----
 utils/tweet_generator.py                      |  10 ++++++++--
 5 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/scripts/extract_and_render.py b/scripts/extract_and_render.py
index 0dcf16d..f8d06bc 100644
--- a/scripts/extract_and_render.py
+++ b/scripts/extract_and_render.py
@@ -39,25 +39,30 @@ def convert_json_to_txt_content(json_path, prefer_original=False):
             # 优先使用原始内容
             title = data.get('original_title', '未找到原始标题')
             content = data.get('original_content', '未找到原始内容')
+            # 优先使用原始标签
+            tags = data.get('original_tags', data.get('tags', '未找到标签'))
             print(f"    - 优先使用原始内容 (prefer_original=True)")
         elif judge_success is True and not prefer_original:
             # 使用审核后的内容
             title = data.get('title', '未找到标题')
             content = data.get('content', '未找到内容')
+            tags = data.get('tags', '未找到标签')
             print(f"    - 使用审核后内容 (judge_success=True)")
         elif 'original_title' in data and 'original_content' in data:
             # 使用原始内容
             title = data.get('original_title', '未找到原始标题')
             content = data.get('original_content', '未找到原始内容')
+            # 优先使用原始标签
+            tags = data.get('original_tags', data.get('tags', '未找到标签'))
             print(f"    - 使用原始内容 (judge_success={judge_success})")
         else:
             # 若无original字段，使用常规字段
             title = data.get('title', '未找到标题')
             content = data.get('content', '未找到内容')
+            tags = data.get('tags', '未找到标签')
             print(f"    - 使用常规内容 (无judge结果)")
         
         # 解决tag/tags字段重复问题，按照修正后的处理逻辑，只使用tags字段
-        tags = data.get('tags', '')
         if not tags and 'tag' in data:
             tags = data.get('tag', '未找到标签')
             print(f"    - 使用tag字段作为标签 (该字段将在后续版本中统一为tags)")
diff --git a/utils/__pycache__/resource_loader.cpython-312.pyc b/utils/__pycache__/resource_loader.cpython-312.pyc
index 815842692f4e8c0d130367eeda53f8b1ec5f2947..dbb3de494603de3007177a64941d0250421dfd1b 100644
GIT binary patch
delta 2115
zcmZ`)YfMvD96zV0_qMmaZK0*_2Ze%w1Lgq7LtZU1K@g^F3aB8V$U`1_TEa9i%%;gq
zTog|>=7-IF*b<E<;`mY%Hv!p%?8`O5xPUP(*|JYtP)JN9%XZF%GRX8K{pJ7v-QRir
zAE%%Br^<9c>vS4GNBZ7`j{Xsy9p;UsxLrXIgx#nSDZnleRGa}pbX~zz05mE`>Q__&
z2*4;1$gQ~T<TR%yqp(Tr5?2)r$)7MkSs(S208So4VQtjkq|8-ZgKF}tB0g5t-PhjU
z+1oy^8fd}2)F3?imRHkRp$p^|+Bs|;HH;Xf#O$%ku^p1PWG<%ksnz*H2V+y7T0Ix_
zE6W-XV}A~iV0$c@F4Uf@{i643!__J&ch?M@2uZH0IkEaTlXYcLMI|po1$5<)GW^@X
z%~-z=O#*?MlNuXCALq|dO{OZDoq-r`69UN|G@0)V*wK`o3wTiLT14bqu~Q=YTUsU~
z%H38+SS8aGy4a_+L7pgEiQgP&R$iw8WCMr8>tQ1oxi}&zH913Q5aEM^s4I3=v=M_`
zbn1{YY>Ivfo5>x{r?p1?E*rUJFp<Bx80eI(1iqEO2(fw)kW#ame93FL7Xk>o$qnAE
zjbESUmBoMXSyuaNZs1at<ocreK}|R*YDWtMcZ5@-T+q1bh?i7v{O^bhD*cB2Q|0|t
z>OJD7uvM#b-+cH1Y5)V!2f|yU%hS5rGNgOWLBCR$ByUtE4(VU59REvw4`y00YXi(`
zbe``F9qsIH!v@C0Fh<?h8w?F}hC8s5aaJ;~X;ic9q#3Z06lzmo90_W@&_vE^--hwz
zjy4CH$uq4DT8Kt0=IvA(`jpg&UUS@q=5x(be10S?;*@qCl-!5rbce~Xc*FeAoH>^N
zD7*YYcKJm6Tz1W(!eMSArMiI4ccu1nt&~|dQ8}?gs`*e#A6^E)?N<<@8#gCy`q(3-
z_day>kpudK;U@Ul7JtQX*&uB#j5I{5CQeEF8l}AlrPM=@Y=<S=;a^N%N%XEPa?soa
zSC)CXeH4Oat$id<QKThOC%OD{V&JjKOojC0%TS2k=58RVQUv#s-l8~d(N{RYbdse)
zkERO(In=qI2`h(1aDSVfIx`wbRWqWDdalB%nlY&u?+_{lG~;ttD$z_aM>Vrbjxw_X
zVzww$=Av1%tuhPEW^vSRWgQkB0yX50`-%49C>R7-F$hEOBox3P>|*UT-fgH?pwuCr
zJmEa#LF@_ML0N1k6-HkItJ^dYZ(+=O=5ac6SwrGva^C3HvJ|hC0l8_cGMcF$Vy*36
z9oRxdQyyQ#Xgk?$T8iKB?u-pYPDYN+IHdR9m+JOQt^;#oJ=q_Z72`$Id7g47`q9ai
zd+Dvn31{5+jkr%Bbdl|rT$oOJEY&frxkFrZFS^sLr4V)!-ufLpKYrU9fcC^8?$srI
z$=*emI%YrX;joUbfY|}@hRaYljmP_l;P6@4K3KjOPp$QP1sqwi{gepZrEnT7@`B)p
z<c#BUuOF`0$Bu|Uk}bJP9*Lz7#L|^T1eqwunXWHGXez-LC`MDoTp1$0POIkh>YoM(
zL-MuKHGbV03%SH0A#5PGV_am`Rad&3ZZ)rN!Qsx<?zZN6*o@_jRZ|+XsAKk)o>Cvj
zeLcs+1zzluUECJx?`E!(9C9bTBdq3g57V=DUQOz2!=S<SOmu)jErWL%>?c3CQyl+=
zhLy-_Xc*FD{Nv8#Kc-~pdy?(>#we#<@@yY#8KeD@KPMKD-#xjyA~fZ+1@h2T9v3Jg
zY4J7r?1{!j3@RAZF=$|Lgu*qjL=j*M1No(4z-l1$AsLH5GVF&-2=K;5g_iTfMF%yX
zDU6&=CS?n^N2jK7t8v)Mhzthub0)8Pl)18!?etPz*wYv6?`|u`9n@v56LM3a*@SQb
Y2n#^B0JIB0{~|`QR{?F$0YiEIzo{J1Bme*a

delta 1778
zcmZuxUrbY17(b`C_qLbzw!OW*rR}A)r3_Y4Pzn@~iR~odL>9r#>9!!F6)=eA-0MGt
zxfmC>WSN?iZI8OdB^XR>BulzCsVp|x%QdTU0hgI<Pm2#kOiW<OcFwI;A9j-S%kTSh
zzVn^)edpxg{w68>A_x_LjG9kfW1cI5A8xr)7xb7w4@jFXfHX5_qE-M7GFJ|*!vOvW
zevd`dR?%KklZ5~&GmK`#qoy0sf<<Nn#~w%HlNwGkKX4Y}re4HPn4RZj-4~mU%__RL
zrRZvXT&X-4pjh{~GIZIo{;UGH{7(VH`yScF%l+5duhr|D-^vZ=26cC`XixnmRj#jC
zh<Q!zAQ8`RPS%23V$#nn0G`l=3I|0$n~Yi(LT1XN48>YLRlzJYiYb-}=2?~qBQMTl
zq;Dn2kh%IbE}GwAT$%;9up>ARB|OIpHsfNb;}HQvwhI5j@;J{*c%9fps;5Jfu*602
zC3g@Haw7EOB<I0rxKGUJ*)+fl(9@Q-&cTu|`jZEmQsMwigEXJ6xM`cCDUi_uW;iWa
z;$+Vd*;B$hEKQ7)<n^VcD-kK%rYp4SGA|bhjcIR`b!oTudVc>}>W%z*tWMjj__cW0
zdI&$THYy9JnFHVyi~{Y=as*P<su_Xo8|S;k*R{SGhh~}oj=u|IGNKoRYy_6LarN-{
zL<TvjV9iXX)l=iz7&21_FaE-I(CQ*I=0qa?+g4XCQ8!`a*rb-3)P^R;Co}4jO&vaZ
zd?GWXj!$Kf8^`UfFociTo1q6^wfmrqbN20=k1SL*{Es~>x^=lR%g>(GJND|*Kv5XP
zTZNl)QEte#Ka6+Zk9Xf0Eynv-OaXZxek!EIh+fm4i{%u3r>0k*C{~@seU5pt`lWV6
zuTDRx%3zN(a&8}d<O}NIrfhq*K6kKSy)&hU1|Iqbb>HA`QdqZ#*H>65?}O{lIHPe*
zyq7et?uloU*|_db6zy$~B$+r7!<wp+q<>7+^hUwmz!rjwHb>7ku;};pG%|~gsf|Rq
z=ZC~_PjOOS|FI!#Hn5THX|vpGwot=1zBkMi%wn&?6l9hxC@3t|gc-`C{4O(7*yQN#
zVhUX>;Ug~^i);@7Po~l9x91htaT6ON%_igP?lYX9Sdb4#rD!BdXv~IaBfTmvs;44-
z@TdllN<og65;^fDX}8-$s7fDsxHOQApGa-o9;%k`M$fAM#dDtBpFNP>otr8|3X<;L
zQ?&QvPPs|&GK+1VZjt#}WV^ldXXUF9Dp;vZz<NAb+2?Z8IBYpQ34W`x6OQ30mAB#P
z`K!JZWPLL%>C!d-5wb25_}p(s0&zz)0aVJ9uo#E-VyhDM(s;<onIr6F%1Nckr4fai
zBk&nm;dnlQFDTyz5=K?GW+PdWbjw3~>wSCc`U(Syxua@RZK|EgM;xhUCg02wK0<?*
z2j#|nPt1Q+)k+iPH78yUe(2jpEKAm5q>Yb|c3*-+$f&?>+#m9}=<hIPMH%(Pgw_#8
z0emtPNm@(A=p#1cxs@vNe+R8V`>E6W6#6N=OW{5Id#En(>hSd4mJUymTx)1U19zS%
z)t~UdhU*T}$@R0zS<=akMf(<P4JU*+lUE$61}5LYrsB9S+~4k{kvl2$P#B=_0fiv~
zH^3@EfMg2BFF=7#bsGN~K4eR*GJtchm~3nU?}*5m!)3>qxpdeVRctD2QHC<ZiH2OW
qb{;*JJ~5Hmjz)<{zgIOtfRqNpWxy{3VHwz-oA#QZ?<t^Yto;w8<e8rU

diff --git a/utils/__pycache__/tweet_generator.cpython-312.pyc b/utils/__pycache__/tweet_generator.cpython-312.pyc
index 982c6ac9160e3acd6c775ba2c6ab54b8487a689c..ea6567f717c5e43e11dc012829a5cac8f31e1d64 100644
GIT binary patch
delta 1781
zcmZvddrVVT9LMkJ?X7)UC=a16Em%8>JPHcpl!C<<FoxiCPR013<)MAh+cFp_P{1H;
zGO=fg(aoeL+rs9S)!k*&O#NpWGN<!0$jln{(hcW^Jak*K{k8KeYD~71^mBfX@9&)6
zbAGq&dslGstU&RlLLsHdI8mL}xbmE0uSS|HARgrxcoT8nc0E2TmEe~(G7SSN7=FQ7
zv|t_DKvAsB$KbHWg!_&0LK{t-Nlm&zvLGO0*(31=YR>>g)sn?K6dj9w2nUSX7#rhP
z*hK9_{W@h^M2d^Gh7{T3;zdssSCK;HLK*JU>X8D!tu17V{qloKJhmq;A?A@5QAewV
z)X`Uj)Ph=QlahrKQ>Im7wwNr<ZUWc=+p%2tAgz;_Bjn>PAkP(`=0%iTH7QmeQw}Kk
zq_q>`BCOTtsAWegyDNu^-lX|u1>>o!czt2m&&O_<5N^c#^t!FeWAXv{P<tpXWIbaD
zC4~0BNAp>haZ@p$UUI&jPpFs>Kl6u(QY{&3!}|M_M6$ZyIw_AkRyk0~YjTD?;n;V)
z!}f4q_=Ru<pIkB_Ux!Ea<!X6!*Hv#Ac%6A%oy#ZZh4p-_bwXImrKfJ8MZNGChq2$_
zRCj?H0CWR-01wmjvQ4O;t5{|cqdwf9wLY<hBqNKPS+}#<=iOW9cGWjEurK1DvKmE@
zVh8b-?470ns2Hc)W_LuFizr#Ur^W8|?e#S|?d}%eT^)!IbA0wT0qV!b+|8&9J99N?
z2oL0%(ICE%`#uVD4SCPgsFk~%Z;lg%pdyGLY%WBXaAsK{^5D8Me@^tvpM%_Tz=r@1
z&<fZE*bS%!e1sL{U80L%Uc$lh=VTpVt^pjybLFSeE$-EdfC%}pvgQa1;uAG}sFKrd
z9gam++?Bc=0(>(=hfVeIs18~OSJ&%g&=1jZ(1tzxH2Agp+W*878#+-#v|_lS_;IZ$
z*jSBfa7|+}8o`Ge>yQlJXski`Tv}5aLMEKnq{jyiPNC~u`Jr!^n2*UWY&_r-z<JK>
z8Ar0{mEDFw4_0`;MkCyh-aMi9Ply-*e-U`e%R{B6vRVlIjkDW!qJQv_wgTA&2z&~-
z$X#huBGk*>e5sj6Qm(8c4r%ij`hGhILI4k76z~OZ?KH@EFvkEaeyejL`3x~5jKke(
zXCt(8e_MoU^4XnU_9_l`oxnK(b651Ld{FU!_w~fnfaGVuFMuh)9l)=EyMQ3zH^A?J
zX}~>z2`~eIHjczO-1}{I8+5A=PX;8Y15X8<=qBFQU5otqT(@5o0r5q=vgZO);d?!W
z5*Z?=wwXk6UhjImZle&plSR0_S1a8`GHb|rdIz~s?{|{;hi~FqzNnI4`z)VaH6h=^
z4Gs<qx>g82i?ilAjLhdkqhp0&B-89tQCBpp2}g^q3YUetDi;u56EAV+Q`huXk=rc1
zZnP?i_f>(FCew|lRfN1e>Wsz6i8Gc`nn<Sai?W*y)DO!`)nvMDDV2DZ3Z@oWm1Me;
zv{6i^u4yCVFiWPFi^=BGE3#Zl=3bf@n4xh=nHecV>=~8BwT7P2B)AIbnF0oEi^QEp
z&x*9}bb2<O0o%-w4QKNu&8hUPC8=3M&uJK9&lw~hJw2C|;Zf7`YKGYJOC_v<na|8-
zb<F(=9YdT9nmq-FTL(BzkU`&B86D##4W~}R__kyrA!HwbXg4`v78<sSjRAc&JBmAw
zr^_-S9{n9&ia$8MBQ=Tm5n-L1wX<TVQ$rs9KUgzZ9DXWjvA|C>3-=hIn_ND&AKWki
onl{4Jlh*J>ga_Pi2YU;`CRmj5wKO~I>@FM%rl3ada<EMBFHrCOP5=M^

delta 1760
zcmZvcdrVVT9LLYOx3~260fiz^Z!1Ma3JRqI1RUUs2+S>0bemDwoP+YRwB@#)Xv#}a
zAwlOnCgPlCE?d@Vmgug_)QFjL)A^WQq**qG+aD~NZOFqJm$)U{`QbDs+qvoIe9!N3
z&*?emclRG<wjW`1S9CfxqV4pGqz2`2-FiWL6(h6{1sR{#k4e8^7Fh5@3o=Kk58B33
zghY*>C0`0wMJ-E(!j$?n_s~fLHID*I(JKLjsu_gZbZwFTf721F<$}6e<rbQ70M$m&
z{4J(rbHwBPh$r&tRBcgoDc4M?m?!s5d2CKlOODNtog36Y3d5+sOo940DbN#BsNuEd
zakX-U^NNjbzvv{bbAC5>&{;0fpD&57rV~vokv3*jrIRD=VbxQlKQXg|?|uFqgKW+m
zHrnKbtYcbPT@+RnPic`+kQK8=4JPmzY<H24%kPWm#l1N+f46$TK2W(odmv??<$Z&k
zvS>IdPqyTrRLhY|!>ZDIN)!{nyOku~MLeJ1>loEW%dx2;97_6#eK!ztgc?HY<e2=h
zw%{IFW~qo%_m=I=mgB9%hBP@QJ!F&Bi^Gbir4KDDSY;1gy||71mE26+^Ne^K>7G}H
z`=rbBid1+9QDrQRdxJ)dM0!O}v)AuiU*mDtHP(xp$cBstCAh?0<a)+hYZpXu&7NAP
zOY8<6B|2LIPM?3hzp>fr3HX1R1Hs#*m6<OxxQ~30RgOE!oh$*rMR>an?;-{EkMVKo
zw*6%mw@SsiHiPmoWDw%3$irvJQA&}FRRl9T!Tbpz9UuWt09pa-09Al$z^7#AvmMGo
zkY`EE^3|Glkc$C3iFf%yJSIgx*Qvw-@@C}@Oi28y?KodLv?>&hOQrJKHB4d+jh$(=
z)xpvO=-}gvi&0R~<Rn9Nc1``Gn5H^1{hC0m_0|8e9jR}}jimBb11W4M#*k+udmB~?
z<q8xmT8)yC9r^Gk2BWnI(`Pr|e}??sP=hfsHwmP@aS=`@nx;f4+{j=&2PC=QbPJ!C
zZf^XR(|<-+AjSet0!~TiTZXY_E9jpA-Q=+E2EHiy{dR@vK3I|ge^B^nMInntG=bp(
zIoG-t|4D?+xth~pI0G1zsy0Vq+#_w=;$^W$8f`ORQ?6JHzLx+qM<s%~1h`Bz9m$#@
zkT(E6QqU2`?~%yPczlJ-@7%|UUTP<Uowl@@lca_H|Ht{{KbYNRPWu%C3E(%tIN%Oo
z0$>G90(wYQml3y<Kvy&Vj*N9x<2Lfl)}ZnSa1N3$ww}TU670_7wX|}5<<v^fb(fIO
zmvO{nR1&_&q^_c##dL3ceUh!`Ca<SOmLumblXJ@CnDVf;LbB`%F&$Rsyuo323EWqT
z6ql8`u)yY4p^HL3<=4dyojV&{)-se=Rq>v5bT!eT^rR@hNpVEc(Y0I$OS|g@F_f})
zW-z4J(!r3qB#w4J7G`*j=%@6O*|ZxgD&boam|GE!DB9glC@rR4$GE_`QuuL;icUUm
zUFhbxJMk)DqToi{1mdWkQ1R|8b|QL~djUJKfCHV)d*-qeg(i=goiuY)Po{8m!pU^r
z8^=y&CwQaTsc4SsDI?!9i=CRA)}m*p^&HjH3A||Jrjs(nSnlrfSdI?TSn&XCEYym4
zh$4->ojucrm(;(#AKLdz9y%1_JqW6%n=8V_dqLDge1>?5==Y>*7J_}|UL})~J!>pR
zIv!EfctoeDf;<8IJg6*CC&;NiMOE+_7U2v>Sok*g!SFZ$F5d`O=kjduk0@UEcwFKb
nnA6~;oIl`oIYl_-hqW*VTu=lLRE6|I7H~O>#YU-;R51Sl1#ss)

diff --git a/utils/output_handler.py b/utils/output_handler.py
index c902004..247d8e3 100644
--- a/utils/output_handler.py
+++ b/utils/output_handler.py
@@ -95,11 +95,14 @@ class FileSystemOutputHandler(OutputHandler):
         import copy
         output_data = copy.deepcopy(content_data)
         
-        # 确保tag和tags字段保持一致
-        if "tag" not in output_data and output_data.get("tags"):
-            output_data["tag"] = output_data["tags"]
-        elif "tags" not in output_data and output_data.get("tag"):
+        # 统一使用tags字段，避免tag和tags重复
+        if "tag" in output_data and "tags" not in output_data:
+            # 只有tag字段存在，复制到tags
             output_data["tags"] = output_data["tag"]
+            del output_data["tag"]
+        elif "tag" in output_data and "tags" in output_data:
+            # 两个字段都存在，保留tags并删除tag
+            del output_data["tag"]
         
         # 确保即使在未启用审核的情况下，字段也保持一致
         if not output_data.get("judged", False):
@@ -108,6 +111,9 @@ class FileSystemOutputHandler(OutputHandler):
             output_data["original_title"] = None
             output_data["original_content"] = None
             output_data["judge_analysis"] = None
+            # 添加original_tags字段
+            if "tags" in output_data and "original_tags" not in output_data:
+                output_data["original_tags"] = output_data["tags"]
         
         # 保存统一格式的article.json
         content_path = os.path.join(variant_dir, "article.json")
diff --git a/utils/tweet_generator.py b/utils/tweet_generator.py
index 85146ed..1c85b45 100644
--- a/utils/tweet_generator.py
+++ b/utils/tweet_generator.py
@@ -518,14 +518,20 @@ content: {content_json.get('content', '')}
                         judged_result = content_judger.judge_content(product_info, content_to_judge)
                         if judged_result and isinstance(judged_result, dict):
                             if "title" in judged_result and "content" in judged_result:
-                                # 使用审核后的内容替换原内容
-                                logging.info(f"  内容审核成功，使用审核后的内容替换原内容")
                                 # 保存原始标题和内容
                                 content_json["original_title"] = content_json.get("title", "")
                                 content_json["original_content"] = content_json.get("content", "")
+                                # 保存原始标签（优先使用tags，如果没有则使用tag）
+                                original_tags = content_json.get("tags", content_json.get("tag", ""))
+                                content_json["original_tags"] = original_tags
                                 # 更新为审核后的内容
                                 content_json["title"] = judged_result["title"]
                                 content_json["content"] = judged_result["content"]
+                                # 保留原始标签，避免重复
+                                content_json["tags"] = original_tags
+                                # 删除可能存在的重复tag字段
+                                if "tag" in content_json:
+                                    del content_json["tag"]
                                 # 添加审核标记
                                 content_json["judged"] = True
                                 # 添加judge_success状态