diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..e08cae4f4e98ccc30348b9eb4a8bef3e7521e137
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,3 @@
+*.json filter=lfs diff=lfs merge=lfs -text
+*.jsonl filter=lfs diff=lfs merge=lfs -text
+*.pdf filter=lfs diff=lfs merge=lfs -text
diff --git a/README.md b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..b05e916a40698a0249e9074e33bcdd22fdb55767
--- /dev/null
+++ b/README.md
@@ -0,0 +1,8 @@
+title: PDF Extraction Comparisson
+emoji: 📉
+colorFrom: green
+colorTo: gray
+sdk: gradio
+sdk_version: 5.27.0
+app_file: extractor_compare.py
+pinned: false
\ No newline at end of file
diff --git a/extraction/non_truncated/record_0.json b/extraction/non_truncated/record_0.json
new file mode 100644
index 0000000000000000000000000000000000000000..4c37ac383293d77cc4441c0ce7382a589c9d4f1c
--- /dev/null
+++ b/extraction/non_truncated/record_0.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:167521da014ee1816adfa9667e045dc5835ebbfde0d4c036a1c354e15f51dbca
+size 63882
diff --git a/extraction/non_truncated/record_1.json b/extraction/non_truncated/record_1.json
new file mode 100644
index 0000000000000000000000000000000000000000..a3ab4694e2e87197f97b67a723d3511783fb5817
--- /dev/null
+++ b/extraction/non_truncated/record_1.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f0a96a2e7dc9833c93b2817f91ea2cdb46844bd8f216b34ec88bbb3b404183e
+size 2355885
diff --git a/extraction/non_truncated/record_10.json b/extraction/non_truncated/record_10.json
new file mode 100644
index 0000000000000000000000000000000000000000..2cd6b56d56a57acc12a5bfe3114e2c251bfe4285
--- /dev/null
+++ b/extraction/non_truncated/record_10.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09b4a9ac963d2795aef7a86d20944049d2554a7cf75dd94a58c17115dc581299
+size 1477192
diff --git a/extraction/non_truncated/record_11.json b/extraction/non_truncated/record_11.json
new file mode 100644
index 0000000000000000000000000000000000000000..44ea6089d174860fd9fb9e7534c8ea11991f227b
--- /dev/null
+++ b/extraction/non_truncated/record_11.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cf21c60728cba6f508dbf44c98cd4b968df2cce2bfcb05f0647762edf54f487
+size 68649
diff --git a/extraction/non_truncated/record_12.json b/extraction/non_truncated/record_12.json
new file mode 100644
index 0000000000000000000000000000000000000000..c4f847ebd730aae5cd082006c772bb013e8fa253
--- /dev/null
+++ b/extraction/non_truncated/record_12.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc2d0f828fdf221ee601cc6904e599665b4ae12082e2b9d2a76c2f7d32528921
+size 58362
diff --git a/extraction/non_truncated/record_13.json b/extraction/non_truncated/record_13.json
new file mode 100644
index 0000000000000000000000000000000000000000..1abe1c27b9d808dcc5bbe1b15154dfed49ba4920
--- /dev/null
+++ b/extraction/non_truncated/record_13.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2182ed09edf5a0aa885f1ac28dddf070d96342c8606fd27e4315e44b39ade036
+size 263549
diff --git a/extraction/non_truncated/record_14.json b/extraction/non_truncated/record_14.json
new file mode 100644
index 0000000000000000000000000000000000000000..ead1dcb8cb4afebf0545d38313c30b40c0efccb9
--- /dev/null
+++ b/extraction/non_truncated/record_14.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3066f1b77f8d4190cf2928a09f883dc6b3f628cf257a9479d9bf822df465e389
+size 610239
diff --git a/extraction/non_truncated/record_15.json b/extraction/non_truncated/record_15.json
new file mode 100644
index 0000000000000000000000000000000000000000..6ad6cb270ea3ff5f40473c950a9a94579eb59e20
--- /dev/null
+++ b/extraction/non_truncated/record_15.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb949316ebfd20294ddd0c1e1ac73a748da4a3f1bfefc819f2c292b54e4ef55
+size 225940
diff --git a/extraction/non_truncated/record_16.json b/extraction/non_truncated/record_16.json
new file mode 100644
index 0000000000000000000000000000000000000000..899f93eb060cd8ada9a57f818ba52a6dd86a0320
--- /dev/null
+++ b/extraction/non_truncated/record_16.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3f7bec9722662ec468459191a19077ccbdda6f4c4bd7156f0d3ba46908afed5
+size 207706
diff --git a/extraction/non_truncated/record_17.json b/extraction/non_truncated/record_17.json
new file mode 100644
index 0000000000000000000000000000000000000000..a66a2c8d33838d582f00c02c8e36424c73369b4d
--- /dev/null
+++ b/extraction/non_truncated/record_17.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c6eeee023c92b02ccc1dcb9d775e80eab03a82f78a062e266590427aa8cd5cd
+size 151205
diff --git a/extraction/non_truncated/record_18.json b/extraction/non_truncated/record_18.json
new file mode 100644
index 0000000000000000000000000000000000000000..a8fcf3a1726a24603d0e68188cc4dcbaefefb003
--- /dev/null
+++ b/extraction/non_truncated/record_18.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca829e8020f472452f7c41ec63f3ddfcdd6dce0b04167c383c311e290129124
+size 730429
diff --git a/extraction/non_truncated/record_19.json b/extraction/non_truncated/record_19.json
new file mode 100644
index 0000000000000000000000000000000000000000..85c2a9f24d3df17481c713e3cf65a1900c194b77
--- /dev/null
+++ b/extraction/non_truncated/record_19.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a5e819288bc5b404228abe38e4e07c966a13597f8cb1ae0336c1d58cebd0573
+size 4347952
diff --git a/extraction/non_truncated/record_2.json b/extraction/non_truncated/record_2.json
new file mode 100644
index 0000000000000000000000000000000000000000..32f98892aaa35ee150b12726d6c23c0f2defadfe
--- /dev/null
+++ b/extraction/non_truncated/record_2.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26cb541e4ed8038eb44dce3f74a40b2a748f09e3ce98168c95b1ae6959699e31
+size 200511
diff --git a/extraction/non_truncated/record_20.json b/extraction/non_truncated/record_20.json
new file mode 100644
index 0000000000000000000000000000000000000000..869a740aaf4a8ff81e9be6632a714e05aa1822dc
--- /dev/null
+++ b/extraction/non_truncated/record_20.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3695a6b19de6b483e6f9f4ce417584d3bca274d1a18bb88f1fc3bc3a4e5a372
+size 34621
diff --git a/extraction/non_truncated/record_21.json b/extraction/non_truncated/record_21.json
new file mode 100644
index 0000000000000000000000000000000000000000..65b518fdc714f22b964c9b9a631d5fd51b4c0f20
--- /dev/null
+++ b/extraction/non_truncated/record_21.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60619c8ebe2caba7c9e2ea3304418a461bd351b90ec6a1a3a3af31fb92ef0196
+size 1403253
diff --git a/extraction/non_truncated/record_22.json b/extraction/non_truncated/record_22.json
new file mode 100644
index 0000000000000000000000000000000000000000..569f6ac1e4e00252fdc63ab6399fc0307bcad136
--- /dev/null
+++ b/extraction/non_truncated/record_22.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9939ba567c0e42fd17ecef79384c5125e56cb5bd53f2229b0115301d048ecc00
+size 505232
diff --git a/extraction/non_truncated/record_23.json b/extraction/non_truncated/record_23.json
new file mode 100644
index 0000000000000000000000000000000000000000..6b337aeb0d4141faf8693c0482a765184900b61e
--- /dev/null
+++ b/extraction/non_truncated/record_23.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81be2481d68dd31f741082419de4dcfd8aa4575200d1fddad407f476a1dedd76
+size 889179
diff --git a/extraction/non_truncated/record_24.json b/extraction/non_truncated/record_24.json
new file mode 100644
index 0000000000000000000000000000000000000000..c441f6383c4364e23132c535c9fedebd1a6fe3b6
--- /dev/null
+++ b/extraction/non_truncated/record_24.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf0f8312ea71e37843d206374692ae41e92f42d442a4c5ec7599eb3ca59e0a88
+size 951925
diff --git a/extraction/non_truncated/record_25.json b/extraction/non_truncated/record_25.json
new file mode 100644
index 0000000000000000000000000000000000000000..5e47396101f03ec11e5040e32c24d3624177bd9f
--- /dev/null
+++ b/extraction/non_truncated/record_25.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86828204d28fb8bbc5031f49548d8a9949767c58ffae242bd401a8996461475b
+size 30260
diff --git a/extraction/non_truncated/record_26.json b/extraction/non_truncated/record_26.json
new file mode 100644
index 0000000000000000000000000000000000000000..36fdea0db923881ec056b179f3891ba908614f90
--- /dev/null
+++ b/extraction/non_truncated/record_26.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a25e63f13b7efa1af614ed705b31a05ea32481e9e354964e1a207652b66a4e19
+size 189677
diff --git a/extraction/non_truncated/record_27.json b/extraction/non_truncated/record_27.json
new file mode 100644
index 0000000000000000000000000000000000000000..5087b222a3f4ea47224d33e2a9c6d4a0e7dc8709
--- /dev/null
+++ b/extraction/non_truncated/record_27.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7709908a0df97a54b8b99f423d646bb280904efa26c97b83989021aacf32cb5
+size 384357
diff --git a/extraction/non_truncated/record_28.json b/extraction/non_truncated/record_28.json
new file mode 100644
index 0000000000000000000000000000000000000000..ab1e166fa0ab73c0fca7ba201cccd368bec11db3
--- /dev/null
+++ b/extraction/non_truncated/record_28.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa2f0b2e981d2972a9c8a52234a5a736eb0f6c86cdb36c61e76b8580a4ec6583
+size 38877
diff --git a/extraction/non_truncated/record_29.json b/extraction/non_truncated/record_29.json
new file mode 100644
index 0000000000000000000000000000000000000000..204ee9091493d885e963e8cd5731b0a9f6eae61c
--- /dev/null
+++ b/extraction/non_truncated/record_29.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ad40b768f55f12542d29a4abb4c832188401046a13611671199350b4c95d921
+size 135635
diff --git a/extraction/non_truncated/record_3.json b/extraction/non_truncated/record_3.json
new file mode 100644
index 0000000000000000000000000000000000000000..dc6b12842d1e0a042a6db92b242801317af223a1
--- /dev/null
+++ b/extraction/non_truncated/record_3.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b1ff68c4d0a3f81aaa60d970767832f823c498286f8c149efeb25b9e6079648
+size 655199
diff --git a/extraction/non_truncated/record_30.json b/extraction/non_truncated/record_30.json
new file mode 100644
index 0000000000000000000000000000000000000000..3b65913c2da0e90f573021294d697b4e9130107b
--- /dev/null
+++ b/extraction/non_truncated/record_30.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5ac40be45da7450a9430f86b45eddb1f0913f0884ed84e67dff6a9d9eae0596
+size 196521
diff --git a/extraction/non_truncated/record_31.json b/extraction/non_truncated/record_31.json
new file mode 100644
index 0000000000000000000000000000000000000000..ede2bc88082f353fa0f0e2f924444bbf40307897
--- /dev/null
+++ b/extraction/non_truncated/record_31.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3120768e0f4f63acfb4ea6c504634c129d40179730e9a268170f4a4cdc1904d
+size 885829
diff --git a/extraction/non_truncated/record_32.json b/extraction/non_truncated/record_32.json
new file mode 100644
index 0000000000000000000000000000000000000000..df7a54e66e36e1a924f976d01ab91985a1d6c68f
--- /dev/null
+++ b/extraction/non_truncated/record_32.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f86e7ab93d134a1fa59bfe062ba6fc226a323f976012320ac61d9b3e56bde0b
+size 372363
diff --git a/extraction/non_truncated/record_33.json b/extraction/non_truncated/record_33.json
new file mode 100644
index 0000000000000000000000000000000000000000..648a4c5d50c3d4d7f07a9ccd6a083deff33ce7b8
--- /dev/null
+++ b/extraction/non_truncated/record_33.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ecf8838485b6667b1362f39f5a0d7886b4a05ed61f80fccdf2f71b9fe9b98c7
+size 5082783
diff --git a/extraction/non_truncated/record_34.json b/extraction/non_truncated/record_34.json
new file mode 100644
index 0000000000000000000000000000000000000000..5bab56f56adfde377c0d245c9e2aa9a8ef05ffbf
--- /dev/null
+++ b/extraction/non_truncated/record_34.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:395237b321ad926362bd520315d527184261f73cf031533aa170d2a36ef80409
+size 38339
diff --git a/extraction/non_truncated/record_35.json b/extraction/non_truncated/record_35.json
new file mode 100644
index 0000000000000000000000000000000000000000..afb5c249c73ac06f0d5c2dcc68f314d8b33e7464
--- /dev/null
+++ b/extraction/non_truncated/record_35.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45172f2b9c3975dc0fa102323d65c98606edaf0dfcfaf1e94c5075bb6ba58eb9
+size 363612
diff --git a/extraction/non_truncated/record_36.json b/extraction/non_truncated/record_36.json
new file mode 100644
index 0000000000000000000000000000000000000000..3803304f084ce370e6e42547251b19ab56a8f328
--- /dev/null
+++ b/extraction/non_truncated/record_36.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1838aa1feed4c9f70dd600276f4d83249ab8f046acc9d67a9028abacde224af1
+size 696226
diff --git a/extraction/non_truncated/record_37.json b/extraction/non_truncated/record_37.json
new file mode 100644
index 0000000000000000000000000000000000000000..58d483d94be6ca4adc364cecb29e576bc032ab54
--- /dev/null
+++ b/extraction/non_truncated/record_37.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eebf8813eba242eaf54427058dadd927c037074645fb6872b068f692e35d8cc
+size 188427
diff --git a/extraction/non_truncated/record_38.json b/extraction/non_truncated/record_38.json
new file mode 100644
index 0000000000000000000000000000000000000000..36e3bde5fbf7c06207f9b25e42d984c38b257ec2
--- /dev/null
+++ b/extraction/non_truncated/record_38.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c90ff6c48476635f96802b94a776605f887bcbc305b23e1b4c579c3f8ae567b9
+size 858917
diff --git a/extraction/non_truncated/record_39.json b/extraction/non_truncated/record_39.json
new file mode 100644
index 0000000000000000000000000000000000000000..a922ee7d115b2e502f6d591e0d1b9c5b7c47b745
--- /dev/null
+++ b/extraction/non_truncated/record_39.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0859d7ced21f7147b79de2e68798076063f3630379e1bd62097750191658d2ba
+size 231498
diff --git a/extraction/non_truncated/record_4.json b/extraction/non_truncated/record_4.json
new file mode 100644
index 0000000000000000000000000000000000000000..d8d3300f172c27069e50cedab82631aef67c5c03
--- /dev/null
+++ b/extraction/non_truncated/record_4.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5dc03dd9e70cbdfbe2f1a2010686f18f984ca5c4d94e6f006de0d4225c43c57
+size 336622
diff --git a/extraction/non_truncated/record_40.json b/extraction/non_truncated/record_40.json
new file mode 100644
index 0000000000000000000000000000000000000000..9338c8f24808b483b6dbb6d53592ee71f43b5f1a
--- /dev/null
+++ b/extraction/non_truncated/record_40.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6831ac023d481844a3368394b5853697f6d1ac5b7ee8042f606c2cc72de28492
+size 189523
diff --git a/extraction/non_truncated/record_41.json b/extraction/non_truncated/record_41.json
new file mode 100644
index 0000000000000000000000000000000000000000..bbe7e854a4a87c047f9ebe5804e975246986767a
--- /dev/null
+++ b/extraction/non_truncated/record_41.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b2c3b9c753feb18018980143409d4a670a2b36f85f271d66e3574297391bb69
+size 1717290
diff --git a/extraction/non_truncated/record_42.json b/extraction/non_truncated/record_42.json
new file mode 100644
index 0000000000000000000000000000000000000000..2e2cacdb145f2cca31ac469368cbcc791992714d
--- /dev/null
+++ b/extraction/non_truncated/record_42.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:103d572230006e3231311feaa9fb18541fbe7347330c195e7e49f8462c131936
+size 504077
diff --git a/extraction/non_truncated/record_43.json b/extraction/non_truncated/record_43.json
new file mode 100644
index 0000000000000000000000000000000000000000..d5bd23e52604cd16c1c8a06e1b926a0ea02501d8
--- /dev/null
+++ b/extraction/non_truncated/record_43.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c26bd2a27e3f7c1cae06a965f4b36333d9eac548238f155a9f22502c1edc9ed8
+size 178433
diff --git a/extraction/non_truncated/record_44.json b/extraction/non_truncated/record_44.json
new file mode 100644
index 0000000000000000000000000000000000000000..fc8238bf1ce9d464f391807891228c425fd21b4a
--- /dev/null
+++ b/extraction/non_truncated/record_44.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3c4e63fe0d91dcc7927005b6bbce23aad1f6a76f438b12334952652bf5811f
+size 578235
diff --git a/extraction/non_truncated/record_45.json b/extraction/non_truncated/record_45.json
new file mode 100644
index 0000000000000000000000000000000000000000..f1e4aea189d610c9c06495a2d7f8edfed5ac4846
--- /dev/null
+++ b/extraction/non_truncated/record_45.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4828368d4df51ee31aac1d332965e597d2ad7e7d69eb12c62547984d333b9c7
+size 65628
diff --git a/extraction/non_truncated/record_46.json b/extraction/non_truncated/record_46.json
new file mode 100644
index 0000000000000000000000000000000000000000..e87043da1736b9ca9e498e0edacc0d57dae280f0
--- /dev/null
+++ b/extraction/non_truncated/record_46.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a5c9af7071dc69a5a2ebc8ba8fa75e55dbcca845215d60cfd82662803f05b4
+size 353066
diff --git a/extraction/non_truncated/record_47.json b/extraction/non_truncated/record_47.json
new file mode 100644
index 0000000000000000000000000000000000000000..06e89f852a5decd81c328e461f96d9a3fa0d546c
--- /dev/null
+++ b/extraction/non_truncated/record_47.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cb7f41c1b0950e0a87846f0b79077eadbff0d70dad758904db03df28b4db761
+size 587671
diff --git a/extraction/non_truncated/record_48.json b/extraction/non_truncated/record_48.json
new file mode 100644
index 0000000000000000000000000000000000000000..26ac45e3097418a3b09c5167e1d848a61546dd4c
--- /dev/null
+++ b/extraction/non_truncated/record_48.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a57ef801045de8bb4bd18e4ba6e9385a97b64b1263e8e2f215b53ef2c3a3785
+size 139104
diff --git a/extraction/non_truncated/record_49.json b/extraction/non_truncated/record_49.json
new file mode 100644
index 0000000000000000000000000000000000000000..5292e76b0e7410ff139b60c43594e29e581b6841
--- /dev/null
+++ b/extraction/non_truncated/record_49.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86703b12aaa2586771ad8140184dedfe719c490bfdeaf1098d7b447d03b5a1e3
+size 424608
diff --git a/extraction/non_truncated/record_5.json b/extraction/non_truncated/record_5.json
new file mode 100644
index 0000000000000000000000000000000000000000..eb458583a61a17fa59bfdfc47005943369df1bca
--- /dev/null
+++ b/extraction/non_truncated/record_5.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e81014ac3d91bf012f645e1732f1292ab417a4bffe50f0a37d3ba5bb99cb0549
+size 463779
diff --git a/extraction/non_truncated/record_6.json b/extraction/non_truncated/record_6.json
new file mode 100644
index 0000000000000000000000000000000000000000..098a587ba50d81238061302da5d8b3efb118a3ac
--- /dev/null
+++ b/extraction/non_truncated/record_6.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c581ce6b5af035a25ba1f0a0a3c3c25bb95bb3a8009c17a09430d9d51153a4e7
+size 723599
diff --git a/extraction/non_truncated/record_7.json b/extraction/non_truncated/record_7.json
new file mode 100644
index 0000000000000000000000000000000000000000..5a304ce4d61d539f1f3b78b2a92d99f0f7394913
--- /dev/null
+++ b/extraction/non_truncated/record_7.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e71c27130c52ca65accba4ca8415144d3aafc3167bafcc0e0736978e192b1745
+size 402046
diff --git a/extraction/non_truncated/record_8.json b/extraction/non_truncated/record_8.json
new file mode 100644
index 0000000000000000000000000000000000000000..d11986d47d42afdd733c010b74362b0a84f203db
--- /dev/null
+++ b/extraction/non_truncated/record_8.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be972a275f06769baf5f75d7ee4e3b047c12a31b0aac6a66170af88a50362cc5
+size 1033561
diff --git a/extraction/non_truncated/record_9.json b/extraction/non_truncated/record_9.json
new file mode 100644
index 0000000000000000000000000000000000000000..6932a93b1323e881c0b7b19164acd37b10cdf8d6
--- /dev/null
+++ b/extraction/non_truncated/record_9.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dce810c5186728b2fad7d8cf94216a547aeb72e7b0a701798ff90c1b572e49fb
+size 760019
diff --git a/extraction/truncated/record_0.json b/extraction/truncated/record_0.json
new file mode 100644
index 0000000000000000000000000000000000000000..a49edcc577f73c477302f7bc3f7160a8126aefc6
--- /dev/null
+++ b/extraction/truncated/record_0.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9f854027e90ea9f907f07175a60050b6b8871147798f32390afb7640188061e
+size 25756717
diff --git a/extraction/truncated/record_1.json b/extraction/truncated/record_1.json
new file mode 100644
index 0000000000000000000000000000000000000000..021789b8d7ed6fc0719fc86c6262c7e9a1ddb452
--- /dev/null
+++ b/extraction/truncated/record_1.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b90300a8372c960618048b0d8c3b7ff093bf644a61033d915b80bd963806cf6
+size 4145981
diff --git a/extraction/truncated/record_10.json b/extraction/truncated/record_10.json
new file mode 100644
index 0000000000000000000000000000000000000000..07aab0e2d65770d8b3310c18a22d79cef08470cf
--- /dev/null
+++ b/extraction/truncated/record_10.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f26172de5fa420716f6d430793fbf89cee0ab660b7a17bd120d310cf4c5f1a3e
+size 4194517
diff --git a/extraction/truncated/record_11.json b/extraction/truncated/record_11.json
new file mode 100644
index 0000000000000000000000000000000000000000..65cc29fa6bad5b1c83e8e1c5f07939dbb9630c00
--- /dev/null
+++ b/extraction/truncated/record_11.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2123eb5aa353531fa34c4b50cfc6d246df72f695a19cc94d4d8f292b089ec94f
+size 2580533
diff --git a/extraction/truncated/record_12.json b/extraction/truncated/record_12.json
new file mode 100644
index 0000000000000000000000000000000000000000..91b29cb2c1fce72d41503ee5a6a2e93305cfe5ca
--- /dev/null
+++ b/extraction/truncated/record_12.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef89883caaef3de85db14ce08255574c7700eb1a13c5d2903a23a06765723c5c
+size 3263274
diff --git a/extraction/truncated/record_13.json b/extraction/truncated/record_13.json
new file mode 100644
index 0000000000000000000000000000000000000000..39b065ae92b559b0b97e22026f73a65c98c98e12
--- /dev/null
+++ b/extraction/truncated/record_13.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ee48fa675c83576ae4f2e700be03814c68b319bcdbba5a79f22a5a49d6107cf
+size 41042075
diff --git a/extraction/truncated/record_14.json b/extraction/truncated/record_14.json
new file mode 100644
index 0000000000000000000000000000000000000000..a3c9f3b375eaf9e5d38bb3c5c92a659b7227bf46
--- /dev/null
+++ b/extraction/truncated/record_14.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d937bf2936be2ffd69f5da617721eda8daefb79795be425fe96ec4123a3a81d
+size 8245268
diff --git a/extraction/truncated/record_15.json b/extraction/truncated/record_15.json
new file mode 100644
index 0000000000000000000000000000000000000000..8c3069bbf4d83ff5b48bd68320b3c46c825ca9c9
--- /dev/null
+++ b/extraction/truncated/record_15.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:031fb3f56bae12cde5cf90b0f8975b4ef4996859e8fa99467f82ed1f8918f2fa
+size 24832502
diff --git a/extraction/truncated/record_16.json b/extraction/truncated/record_16.json
new file mode 100644
index 0000000000000000000000000000000000000000..61a3b467a480054ee86040d4030d91dc54448127
--- /dev/null
+++ b/extraction/truncated/record_16.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70a8d663ea0c7e1f9abb740e04b825d5ede2dfc6a42d5b462ca2292b59aa6e72
+size 10585522
diff --git a/extraction/truncated/record_17.json b/extraction/truncated/record_17.json
new file mode 100644
index 0000000000000000000000000000000000000000..e7b985cc9f96b48ca2579725967267d463fce1b0
--- /dev/null
+++ b/extraction/truncated/record_17.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:213bb6081b74e6a5e7317b5de229105a1225cb79e7205e99ef528a3f00f1ee1e
+size 2200094
diff --git a/extraction/truncated/record_18.json b/extraction/truncated/record_18.json
new file mode 100644
index 0000000000000000000000000000000000000000..198838d0c98777d24110b032e88ecf1b7e446c5b
--- /dev/null
+++ b/extraction/truncated/record_18.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9670772adf0fa0660dbf0432cda18327a6a6483ad5e05d1b14292c87ebe33628
+size 8938751
diff --git a/extraction/truncated/record_19.json b/extraction/truncated/record_19.json
new file mode 100644
index 0000000000000000000000000000000000000000..5daece9f6c55753b7eceb010395d760de68689b8
--- /dev/null
+++ b/extraction/truncated/record_19.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c0e0099a03748dc387c75c1754aa1f916449623f83b1c6b83b425a2d6c0284e
+size 2321802
diff --git a/extraction/truncated/record_2.json b/extraction/truncated/record_2.json
new file mode 100644
index 0000000000000000000000000000000000000000..572a1b595cd8f2b948e93cf94bbb4c3a73886b4d
--- /dev/null
+++ b/extraction/truncated/record_2.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0673f4d8c977afee6f5e0c22f6217cf9285552a4424958d79eb6d51e1fdd12a
+size 7181245
diff --git a/extraction/truncated/record_20.json b/extraction/truncated/record_20.json
new file mode 100644
index 0000000000000000000000000000000000000000..3178822925f59c2a090769c56a549377fe76c6df
--- /dev/null
+++ b/extraction/truncated/record_20.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:991245b6596640db34394f15c8a7ae16239707d9ceb175a9820d0dae21e23857
+size 3105031
diff --git a/extraction/truncated/record_21.json b/extraction/truncated/record_21.json
new file mode 100644
index 0000000000000000000000000000000000000000..55cd682c41d5bf3d970bd60d5a966c81141dd3b8
--- /dev/null
+++ b/extraction/truncated/record_21.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2447dc54a55ab851c0c484c3357126000849aaf4155e28aa6afcdaebd1e720bf
+size 10944540
diff --git a/extraction/truncated/record_22.json b/extraction/truncated/record_22.json
new file mode 100644
index 0000000000000000000000000000000000000000..e22920acdfb280cf817bd041ed7dbab4cdeb0ef5
--- /dev/null
+++ b/extraction/truncated/record_22.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75187a3be16a8dc53e8fb28212f4e577d46ba6e052ca9df3a20c0eab5a91cbdb
+size 2416047
diff --git a/extraction/truncated/record_23.json b/extraction/truncated/record_23.json
new file mode 100644
index 0000000000000000000000000000000000000000..af17db423338e9eaf2e1f76c86bb5de0e3acd51f
--- /dev/null
+++ b/extraction/truncated/record_23.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad2d636f829de3d4bb72672441ff7a7a100b632ebca8476e5ad487b174a7b0c9
+size 4818556
diff --git a/extraction/truncated/record_24.json b/extraction/truncated/record_24.json
new file mode 100644
index 0000000000000000000000000000000000000000..fa68211ee093ed5ef705b5dd4272a6d64789cf0a
--- /dev/null
+++ b/extraction/truncated/record_24.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a18111223e430dc7ff325275c99e1d7ebfb67996879772d6f67616a2ebc7eb66
+size 1751628
diff --git a/extraction/truncated/record_25.json b/extraction/truncated/record_25.json
new file mode 100644
index 0000000000000000000000000000000000000000..5c8133e6a28e363b16d0c9b08ab4c801a3e5f775
--- /dev/null
+++ b/extraction/truncated/record_25.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:127a04149e2f7ac4116900d1a7121f83ec7c2f447d59903ec5bb2a93524b6f8e
+size 17538244
diff --git a/extraction/truncated/record_26.json b/extraction/truncated/record_26.json
new file mode 100644
index 0000000000000000000000000000000000000000..f2cd460905e7b907b43036ec7fc24aac9d73480a
--- /dev/null
+++ b/extraction/truncated/record_26.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:848c0b5224b721d57af7f55e29a7962df499b2ef12e91581558d0d9423361d06
+size 3356738
diff --git a/extraction/truncated/record_27.json b/extraction/truncated/record_27.json
new file mode 100644
index 0000000000000000000000000000000000000000..4cc742daac06da161c51287d77b95a3e67d745da
--- /dev/null
+++ b/extraction/truncated/record_27.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dda37e766399267a1440fa7cc935147e87b1aef03e3a9ed4404628fbfed5e85
+size 3583167
diff --git a/extraction/truncated/record_28.json b/extraction/truncated/record_28.json
new file mode 100644
index 0000000000000000000000000000000000000000..e46fbdd448cd9be649a68bc97fc5ad6c43dbd101
--- /dev/null
+++ b/extraction/truncated/record_28.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8421b5ca38cd0ffc41dd6bd5fbc68c378f62bd8cd6b683696f5577d570c14c7
+size 4302378
diff --git a/extraction/truncated/record_29.json b/extraction/truncated/record_29.json
new file mode 100644
index 0000000000000000000000000000000000000000..ff97c1b9639b3bb4920a7db57275884c29c00ea4
--- /dev/null
+++ b/extraction/truncated/record_29.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ccacb911774371f5baca43b6a6f12c8c3146582fc30e6dc29f13cebee3f32dc
+size 3582508
diff --git a/extraction/truncated/record_3.json b/extraction/truncated/record_3.json
new file mode 100644
index 0000000000000000000000000000000000000000..9f166c1a0fff2d57710cb25e98c78b02c0eb3e8b
--- /dev/null
+++ b/extraction/truncated/record_3.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8496fefb4aae23d27ed8264b406b3538c398bf0a64a20a44eb717bc7bc146489
+size 9584404
diff --git a/extraction/truncated/record_30.json b/extraction/truncated/record_30.json
new file mode 100644
index 0000000000000000000000000000000000000000..390bc8e4ada4c3b23cfc029faad0eb4bbb641a71
--- /dev/null
+++ b/extraction/truncated/record_30.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01c03b30a9fe4758d4784cdfbbbfb21f772cab1de362ba84bc2fbbe924f7a916
+size 1723861
diff --git a/extraction/truncated/record_31.json b/extraction/truncated/record_31.json
new file mode 100644
index 0000000000000000000000000000000000000000..9544c20390699d5753e1b4e11dda3a97ef5e29da
--- /dev/null
+++ b/extraction/truncated/record_31.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb9c6baae05618eb62f2b7db923a78f93d657ba8845ffc1d95afe27e7f3afb6b
+size 21088951
diff --git a/extraction/truncated/record_32.json b/extraction/truncated/record_32.json
new file mode 100644
index 0000000000000000000000000000000000000000..27279f2c3d75968aea88a06b3ee3cfed29c05aa8
--- /dev/null
+++ b/extraction/truncated/record_32.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db93f7c4b60c496e2ff10413e0dceddf5b45db0a3c0515f9f73c8ccea92166e0
+size 5262069
diff --git a/extraction/truncated/record_33.json b/extraction/truncated/record_33.json
new file mode 100644
index 0000000000000000000000000000000000000000..9c9d5f0add60b4f31174d84f1749f4756d0e3ce2
--- /dev/null
+++ b/extraction/truncated/record_33.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:151558fc315cc47bf8d9fa010a65a142970f425460a43933288f2871072150bf
+size 59458095
diff --git a/extraction/truncated/record_34.json b/extraction/truncated/record_34.json
new file mode 100644
index 0000000000000000000000000000000000000000..7f93a224d17ec6d4054676650b8fc3f32a4dd216
--- /dev/null
+++ b/extraction/truncated/record_34.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:571dc26a89b61440bedd04e2b47feff4fa41203b6da252c75b40d957d4d02ed1
+size 7331042
diff --git a/extraction/truncated/record_35.json b/extraction/truncated/record_35.json
new file mode 100644
index 0000000000000000000000000000000000000000..ede6387e89915a1cbb7cd3ad8e9db68ee6ee1b53
--- /dev/null
+++ b/extraction/truncated/record_35.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e305511201d9892214e0f1c35552b343bb87b6d018b25c61e3181f5fdd50496
+size 1255709
diff --git a/extraction/truncated/record_36.json b/extraction/truncated/record_36.json
new file mode 100644
index 0000000000000000000000000000000000000000..4abdd5ac39066b5e376712b4ad64bf817f94edf4
--- /dev/null
+++ b/extraction/truncated/record_36.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ff727e3a74f32bbcdd7df4a8fc54fa05cbee106af90ee41e33968c6b42f16a0
+size 9710246
diff --git a/extraction/truncated/record_37.json b/extraction/truncated/record_37.json
new file mode 100644
index 0000000000000000000000000000000000000000..e02074a18b9efaf1442b2737461059752d03572e
--- /dev/null
+++ b/extraction/truncated/record_37.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62c302266f1792d3ad133d71ec6122648db66f2d4c5780a8f0dfe58193d4bafd
+size 1595130
diff --git a/extraction/truncated/record_38.json b/extraction/truncated/record_38.json
new file mode 100644
index 0000000000000000000000000000000000000000..41fa480378c99df1ee48ad15471450f555f80886
--- /dev/null
+++ b/extraction/truncated/record_38.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:287acb8b7e6beb5cd79dba63fa5908a60ed19f227d12c1bd073cc63493165b3d
+size 2124023
diff --git a/extraction/truncated/record_39.json b/extraction/truncated/record_39.json
new file mode 100644
index 0000000000000000000000000000000000000000..c677d7e17822a2eeb490ffa0968195cd5e9e4512
--- /dev/null
+++ b/extraction/truncated/record_39.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fc6886948acfa477ab9c4610a1ef021711d66ad5192c028f069369c671b7cf4
+size 8402211
diff --git a/extraction/truncated/record_4.json b/extraction/truncated/record_4.json
new file mode 100644
index 0000000000000000000000000000000000000000..483a545e8385e77c81f2a9ba8120abeb62c4b9af
--- /dev/null
+++ b/extraction/truncated/record_4.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:687e384c43cdd0c984b2ded2d5be8242ff93e82e1316ab4ad1da6a438ebb5563
+size 6800433
diff --git a/extraction/truncated/record_5.json b/extraction/truncated/record_5.json
new file mode 100644
index 0000000000000000000000000000000000000000..ea9a97fc96cdb6883c3876a80c969e60381ded6a
--- /dev/null
+++ b/extraction/truncated/record_5.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce6c1802678e0302deb7d3833380311ed671c1408f65f539bdd5910607e08bd
+size 4264032
diff --git a/extraction/truncated/record_6.json b/extraction/truncated/record_6.json
new file mode 100644
index 0000000000000000000000000000000000000000..90c4104bfc88a5cc9745fb66d55cb5a7d1ed3aa4
--- /dev/null
+++ b/extraction/truncated/record_6.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:536f130416453676aa8e3f517ce47a66e6ef95c66f2b58931b3c073ed2e6742d
+size 20471457
diff --git a/extraction/truncated/record_7.json b/extraction/truncated/record_7.json
new file mode 100644
index 0000000000000000000000000000000000000000..e7b28ca4cf565f5304ed8b1b77179a5940971995
--- /dev/null
+++ b/extraction/truncated/record_7.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2501caa9a61be302e7ec61b79cd6c6f70c443777175e227190e9b66ca55ba751
+size 13878491
diff --git a/extraction/truncated/record_8.json b/extraction/truncated/record_8.json
new file mode 100644
index 0000000000000000000000000000000000000000..3997ac0f0ed1cd550172e84e01b6b2c4a631d15a
--- /dev/null
+++ b/extraction/truncated/record_8.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47dfa5fceafd678dde151979cc4f9e38bb2cee5bafad2c21a8fb5a3c04f7734d
+size 3308505
diff --git a/extraction/truncated/record_9.json b/extraction/truncated/record_9.json
new file mode 100644
index 0000000000000000000000000000000000000000..262b0789a11c537ab76e76deeb6f9387060d22fa
--- /dev/null
+++ b/extraction/truncated/record_9.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5516b04c9ba03f923229fff21748d1966546ce39110efc475ea173cfc8f06fa
+size 11377676
diff --git a/extractor_compare.py b/extractor_compare.py
new file mode 100644
index 0000000000000000000000000000000000000000..5e905a71b4a505e541cbe356709b20c99e69d802
--- /dev/null
+++ b/extractor_compare.py
@@ -0,0 +1,481 @@
+import gradio as gr
+import os
+import json
+import base64
+import tempfile
+from pathlib import Path
+
+EXTRACTORS = ['pdf_plumber', 'py_pdf', 'docling', 'extractous', 'pypdfium2', 'pymupdf', 'pymupdf_llm']
+
+def add_page_breaks(text, page_offsets):
+    """Add page break markers to text based on page_offsets."""
+    if not page_offsets:
+        return text
+    
+    result = []
+    last_offset = 0
+    for offset in page_offsets:
+        result.append(text[last_offset:offset])
+        result.append("\n<---page-break--->\n")
+        last_offset = offset
+    
+    # Add any remaining text
+    if last_offset < len(text):
+        result.append(text[last_offset:])
+    
+    return "".join(result)
+
+class ExtractorComparer:
+    def __init__(self):
+        self.json_files = []
+        self.current_index = 0
+        self.current_data = None
+        self.temp_pdf_path = None
+        self.current_pdf_bytes = None
+    
+    def load_files(self, directory_path):
+        """Load all JSON files from the specified directory."""
+        self.json_files = []
+        try:
+            for filename in os.listdir(directory_path):
+                if filename.endswith('.json') or filename.endswith('.jsonl'):
+                    self.json_files.append(os.path.join(directory_path, filename))
+            
+            if self.json_files:
+                self.current_index = 0
+                file_progress, annotation_status = self.get_progress_info()
+                return file_progress, annotation_status
+            else:
+                return "No JSON files found", "No files loaded"
+        except Exception as e:
+            return f"Error loading files: {str(e)}", "Error"
+    
+    def load_current_file(self):
+        """Load the current JSON file data."""
+        if not self.json_files:
+            return None, "N/A", "N/A"
+        
+        try:
+            with open(self.json_files[self.current_index], 'r') as f:
+                self.current_data = json.load(f)
+            
+            # Extract PDF bytes from pdf_plumber
+            pdf_bytes = None
+            debug_info = ""
+            if 'pdf_plumber' in self.current_data:
+                plumber_data = self.current_data['pdf_plumber']
+                if 'media' in plumber_data and plumber_data['media'] and isinstance(plumber_data['media'], list) and len(plumber_data['media']) > 0:
+                    media_item = plumber_data['media'][0]
+                    if 'media_bytes' in media_item and media_item['media_bytes']:
+                        try:
+                            pdf_bytes = base64.b64decode(media_item['media_bytes'])
+                            self.current_pdf_bytes = pdf_bytes
+                        except Exception as e:
+                            debug_info = f"Error decoding media_bytes: {str(e)}"
+            
+            # Create temporary file for the PDF if we have bytes
+            if pdf_bytes:
+                if self.temp_pdf_path:
+                    try:
+                        os.remove(self.temp_pdf_path)
+                    except:
+                        pass
+                
+                with tempfile.NamedTemporaryFile(delete=False, suffix='.pdf') as temp_file:
+                    temp_file.write(pdf_bytes)
+                    self.temp_pdf_path = temp_file.name
+                
+                # Convert to base64 for passing to the frontend
+                base64_pdf = base64.b64encode(pdf_bytes).decode('utf-8')
+                
+                # Generate progress information
+                file_progress, annotation_status = self.get_progress_info()
+                
+                return base64_pdf, file_progress, annotation_status
+            else:
+                file_progress, annotation_status = self.get_progress_info()
+                return None, file_progress, annotation_status
+        except Exception as e:
+            return None, "Error loading file", "No annotation"
+    
+    def get_progress_info(self):
+        """Generate progress information and annotation status."""
+        if not self.json_files:
+            return "No files loaded", "No annotation"
+        
+        current_file = self.json_files[self.current_index]
+        filename = Path(current_file).name
+        
+        # File progress information
+        file_progress = f"File {self.current_index + 1} of {len(self.json_files)}: {filename}"
+        
+        # Check if this file has been annotated with a best extractor
+        best_extractor_file = os.path.splitext(current_file)[0] + "_best.txt"
+        annotation_status = "Not annotated"
+        
+        if os.path.exists(best_extractor_file):
+            try:
+                with open(best_extractor_file, 'r') as f:
+                    best_extractor = f.read().strip()
+                annotation_status = f"Best extractor: {best_extractor}"
+            except:
+                pass
+                
+        # Count total annotated files
+        annotated_count = 0
+        for json_file in self.json_files:
+            best_file = os.path.splitext(json_file)[0] + "_best.txt"
+            if os.path.exists(best_file):
+                annotated_count += 1
+                
+        file_progress = f"{file_progress} (Annotated: {annotated_count}/{len(self.json_files)})"
+        
+        return file_progress, annotation_status
+    
+    def get_extractor_text(self, extractor_name):
+        """Get text with page breaks for the specified extractor."""
+        if not self.current_data or extractor_name not in self.current_data:
+            return ""
+        
+        extractor_data = self.current_data[extractor_name]
+        if 'text' not in extractor_data:
+            return f"No text found for {extractor_name}"
+        
+        text = extractor_data.get('text', '')
+        
+        # Get page offsets
+        page_offsets = []
+        if 'media' in extractor_data and extractor_data['media'] and len(extractor_data['media']) > 0:
+            media_item = extractor_data['media'][0]
+            if 'metadata' in media_item and 'pdf_metadata' in media_item['metadata'] and 'page_offsets' in media_item['metadata']['pdf_metadata']:
+                page_offsets = media_item['metadata']['pdf_metadata']['page_offsets']
+        
+        return add_page_breaks(text, page_offsets)
+    
+    def next_pdf(self):
+        """Load the next PDF in the list."""
+        if not self.json_files:
+            return None, "N/A", "N/A"
+        
+        self.current_index = (self.current_index + 1) % len(self.json_files)
+        return self.load_current_file()
+    
+    def prev_pdf(self):
+        """Load the previous PDF in the list."""
+        if not self.json_files:
+            return None, "N/A", "N/A"
+        
+        self.current_index = (self.current_index - 1) % len(self.json_files)
+        return self.load_current_file()
+    
+    def set_best_extractor(self, extractor_name):
+        """Record that this extractor is the best for the current file."""
+        if not self.json_files or not self.current_data:
+            return "N/A", "N/A"
+        
+        try:
+            # Create a record about the best extractor
+            result_file = os.path.splitext(self.json_files[self.current_index])[0] + "_best.txt"
+            with open(result_file, 'w') as f:
+                f.write(extractor_name)
+            
+            # Get updated progress info after annotation
+            file_progress, annotation_status = self.get_progress_info()
+            
+            return file_progress, annotation_status
+        except Exception as e:
+            return "Error saving annotation", "No annotation"
+
+def create_interface():
+    comparer = ExtractorComparer()
+    
+    # Custom CSS for basic font in text areas
+    custom_css = """
+    .extraction-text textarea {
+        font-family: Arial, Helvetica, sans-serif !important;
+        font-size: 14px !important;
+        line-height: 1.5 !important;
+    }
+    """
+    
+    with gr.Blocks(title="PDF Extractor Comparer", theme="soft", css=custom_css) as demo:
+        gr.Markdown("## PDF Extractor Comparer")
+        
+        with gr.Row():
+            directory_input = gr.Textbox(
+                label="Path to JSON Directory",
+                placeholder="e.g., /path/to/your/json/files"
+            )
+            load_button = gr.Button("Load PDFs", variant="primary")
+        
+        # Main layout: PDF viewer on left, status and controls on right
+        with gr.Row():
+            # Left column: PDF viewer
+            with gr.Column(scale=3):
+                # PDF viewer using iframe with JavaScript handling
+                pdf_viewer_html = gr.HTML(
+                    label="PDF Document",
+                    value='''
+                    <div style="width:100%; height:700px; position:relative; border:1px solid #ddd;">
+                        <iframe id="pdf-iframe" width="100%" height="100%" style="border:none;" src="about:blank"></iframe>
+                        <div id="pdf-fallback" style="position:absolute; top:0; left:0; width:100%; height:100%; 
+                             display:flex; align-items:center; justify-content:center; padding:20px; text-align:center;">
+                            Click "Load PDFs" to start viewing documents.
+                        </div>
+                    </div>
+                    '''
+                )
+                # Hidden component to store the Base64 PDF data
+                pdf_data_hidden = gr.Textbox(visible=False, elem_id="pdf_base64_data")
+            
+            # Right column: Progress and controls
+            with gr.Column(scale=1):
+                # Progress information
+                file_progress_output = gr.Textbox(label="File Progress", interactive=False)
+                annotation_status_output = gr.Textbox(label="Annotation Status", interactive=False)
+                
+                # Navigation
+                with gr.Row():
+                    prev_button = gr.Button("⬅️ Previous", elem_id="prev_button")
+                    next_button = gr.Button("Next ➡️", elem_id="next_button")
+                
+                # Best extractor selection
+                gr.Markdown("### Select Best Extractor")
+                extractor_buttons = []
+                for extractor in EXTRACTORS:
+                    button = gr.Button(extractor, variant="secondary")
+                    extractor_buttons.append(button)
+                    button.click(
+                        comparer.set_best_extractor, 
+                        inputs=[gr.Textbox(value=extractor, visible=False)],
+                        outputs=[file_progress_output, annotation_status_output]
+                    )
+        
+        # Extractors section below the PDF
+        gr.Markdown("### Extractor Comparison")
+        
+        # Extractor dropdowns
+        with gr.Row():
+            extractor1_dropdown = gr.Dropdown(
+                choices=EXTRACTORS, 
+                label="Extractor 1",
+                value=EXTRACTORS[0] if EXTRACTORS else None
+            )
+            extractor2_dropdown = gr.Dropdown(
+                choices=EXTRACTORS, 
+                label="Extractor 2",
+                value=EXTRACTORS[1] if len(EXTRACTORS) > 1 else EXTRACTORS[0] if EXTRACTORS else None
+            )
+        
+        # Extractor text outputs with applied class for styling
+        with gr.Row():
+            extractor1_text = gr.Textbox(
+                label="Extractor 1 Output", 
+                lines=15,
+                elem_classes=["extraction-text"]
+            )
+            extractor2_text = gr.Textbox(
+                label="Extractor 2 Output", 
+                lines=15,
+                elem_classes=["extraction-text"]
+            )
+        
+        # Event handlers
+        load_button.click(
+            comparer.load_files, 
+            inputs=[directory_input],
+            outputs=[file_progress_output, annotation_status_output]
+        ).then(
+            comparer.load_current_file,
+            outputs=[pdf_data_hidden, file_progress_output, annotation_status_output]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor1_dropdown],
+            outputs=[extractor1_text]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor2_dropdown],
+            outputs=[extractor2_text]
+        )
+        
+        prev_button.click(
+            comparer.prev_pdf,
+            outputs=[pdf_data_hidden, file_progress_output, annotation_status_output]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor1_dropdown],
+            outputs=[extractor1_text]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor2_dropdown],
+            outputs=[extractor2_text]
+        )
+        
+        next_button.click(
+            comparer.next_pdf,
+            outputs=[pdf_data_hidden, file_progress_output, annotation_status_output]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor1_dropdown],
+            outputs=[extractor1_text]
+        ).then(
+            comparer.get_extractor_text,
+            inputs=[extractor2_dropdown],
+            outputs=[extractor2_text]
+        )
+        
+        extractor1_dropdown.change(
+            comparer.get_extractor_text,
+            inputs=[extractor1_dropdown],
+            outputs=[extractor1_text]
+        )
+        
+        extractor2_dropdown.change(
+            comparer.get_extractor_text,
+            inputs=[extractor2_dropdown],
+            outputs=[extractor2_text]
+        )
+        
+        # Add JavaScript to handle PDF display
+        demo.load(
+            None, None, None,
+            js="""
+            () => {
+                console.log('Setting up PDF Blob handler...');
+                window.currentPdfBlobUrl = null; // Store previous blob url globally
+
+                function displayPdfBlob(base64Data) {
+                    console.log('displayPdfBlob called - data length:', base64Data ? base64Data.length : 0);
+                    const iframe = document.getElementById('pdf-iframe');
+                    const fallbackDiv = document.getElementById('pdf-fallback');
+                    
+                    if (!iframe) {
+                        console.error('PDF iframe not found!');
+                        return;
+                    }
+
+                    // Revoke previous Blob URL to free memory
+                    if (window.currentPdfBlobUrl) {
+                        console.log('Revoking previous Blob URL:', window.currentPdfBlobUrl);
+                        URL.revokeObjectURL(window.currentPdfBlobUrl);
+                        window.currentPdfBlobUrl = null;
+                    }
+
+                    if (base64Data && base64Data.length > 100) { // Ensure there's actual content
+                        try {
+                            // Hide fallback message
+                            if (fallbackDiv) fallbackDiv.style.display = 'none';
+                            
+                            // Decode Base64
+                            const byteCharacters = atob(base64Data);
+                            console.log('Base64 decoded successfully, length:', byteCharacters.length);
+                            const byteNumbers = new Array(byteCharacters.length);
+                            for (let i = 0; i < byteCharacters.length; i++) {
+                                byteNumbers[i] = byteCharacters.charCodeAt(i);
+                            }
+                            const byteArray = new Uint8Array(byteNumbers);
+                            
+                            // Create Blob and URL
+                            const blob = new Blob([byteArray], {type: 'application/pdf'});
+                            window.currentPdfBlobUrl = URL.createObjectURL(blob);
+                            console.log('Created new Blob URL:', window.currentPdfBlobUrl);
+                            
+                            // Update iframe source
+                            iframe.src = window.currentPdfBlobUrl;
+                            console.log('Iframe src updated to Blob URL');
+                        } catch (e) {
+                            console.error('Error processing Base64 data or creating Blob URL:', e);
+                            if (fallbackDiv) {
+                                fallbackDiv.innerHTML = '<div style="color:red;">Error loading PDF: ' + e.message + '</div>';
+                                fallbackDiv.style.display = 'flex';
+                            }
+                            iframe.src = 'about:blank'; // Clear iframe on error
+                        }
+                    } else {
+                        console.log('No valid Base64 data provided.');
+                        if (fallbackDiv) {
+                            fallbackDiv.innerHTML = '<div>No PDF loaded yet. Use the "Load PDFs" button.</div>';
+                            fallbackDiv.style.display = 'flex';
+                        }
+                        iframe.src = 'about:blank'; // Clear iframe if no data
+                    }
+                }
+                
+                // MutationObserver to watch the hidden Textbox
+                const targetNode = document.getElementById('pdf_base64_data');
+                if (targetNode) {
+                    // Find the actual textarea inside the Gradio component structure
+                    const hiddenTextArea = targetNode.querySelector('textarea');
+                    if(hiddenTextArea){
+                        console.log('Found hidden textarea to observe.');
+                        const observerConfig = { characterData: true, childList: true, subtree: true, attributes: true }; // Watch for all changes
+                        
+                        const observer = new MutationObserver(function(mutationsList) {
+                            console.log('Mutation detected, checking textarea value');
+                            if (hiddenTextArea.value && hiddenTextArea.value.length > 100) {
+                                console.log('Valid value found in textarea, displaying PDF');
+                                displayPdfBlob(hiddenTextArea.value);
+                            }
+                        });
+                        
+                        observer.observe(targetNode, observerConfig);
+                        console.log('MutationObserver attached');
+
+                        // Try to display any initial value
+                        setTimeout(() => {
+                            if(hiddenTextArea.value && hiddenTextArea.value.length > 100) { 
+                                console.log('Initial value found in textarea, displaying PDF');
+                                displayPdfBlob(hiddenTextArea.value); 
+                            }
+                        }, 1000);
+
+                    } else {
+                        console.error('Could not find the textarea within #pdf_base64_data!');
+                    }
+                } else {
+                    console.error('Hidden data element #pdf_base64_data not found!');
+                }
+                
+                // Add keyboard shortcuts like in app.py
+                document.addEventListener('keydown', function(event) {
+                    if (event.target.tagName === 'INPUT' || event.target.tagName === 'TEXTAREA') {
+                        return; // Ignore inputs
+                    }
+                    let targetButtonId = null;
+                    const key = event.key;
+                    
+                    if (key === 'ArrowLeft') targetButtonId = 'prev_button';
+                    else if (key === 'ArrowRight') targetButtonId = 'next_button';
+
+                    if (targetButtonId) {
+                        const targetButton = document.getElementById(targetButtonId);
+                        if (targetButton) {
+                             event.preventDefault(); 
+                             targetButton.click(); 
+                        }
+                    } 
+                });
+                console.log('Keydown listener added.');
+                
+                // Additional style for basic font
+                const additionalStyle = document.createElement('style');
+                additionalStyle.textContent = `
+                    .extraction-text textarea {
+                        font-family: Arial, Helvetica, sans-serif !important;
+                        font-size: 14px !important;
+                    }
+                `;
+                document.head.appendChild(additionalStyle);
+            }
+            """
+        )
+    
+    return demo
+
+if __name__ == "__main__":
+    demo = create_interface()
+    demo.launch(
+        server_name="127.0.0.1",
+        server_port=7865,
+        inbrowser=True
+    ) 
\ No newline at end of file