Skip to content
Toggle navigation
Toggle navigation
This project
Loading...
Sign in
周伟奇
/
bmw-ocr
Go to a project
Toggle navigation
Toggle navigation pinning
Projects
Groups
Snippets
Help
Project
Activity
Repository
Graphs
Network
Create a new issue
Commits
Issue Boards
Files
Commits
Network
Compare
Branches
Tags
50017fdb
authored
2020-11-10 11:53:10 +0800
by
周伟奇
Browse Files
Options
Browse Files
Tag
Download
Email Patches
Plain Diff
add split \n
1 parent
95966861
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
29 additions
and
3 deletions
src/apps/doc/consts.py
src/apps/doc/ocr/wb.py
src/apps/doc/consts.py
View file @
50017fd
...
...
@@ -896,6 +896,7 @@ OTHER_CLASSIFY_SET = {OTHER_CLASSIFY}
LICENSE_CLASSIFY_SET_1
=
{
IC_CLASSIFY
,
VAT_CLASSIFY
,
MVC_CLASSIFY
,
MVI_CLASSIFY
}
LICENSE_CLASSIFY_SET_2
=
{
BL_CLASSIFY
,
UCI_CLASSIFY
,
EEP_CLASSIFY
,
DL_CLASSIFY
,
PP_CLASSIFY
,
BC_CLASSIFY
}
NYYH_CLASSIFY
=
17
MS_CLASSIFY
=
21
MS_ERROR_COL
=
(
5
,
6
)
WECHART_CLASSIFY
=
12
...
...
src/apps/doc/ocr/wb.py
View file @
50017fd
...
...
@@ -312,7 +312,7 @@ class BSWorkbook(Workbook):
return
res_str
@staticmethod
def
wechart_func
(
row_value
,
pre_col
,
next_col
):
def
rm_cn_char
(
row_value
,
pre_col
,
next_col
):
row_value
=
list
(
row_value
)
if
isinstance
(
row_value
[
pre_col
],
str
):
cn_chars
=
re
.
findall
(
consts
.
CN_RE
,
row_value
[
pre_col
])
...
...
@@ -324,6 +324,29 @@ class BSWorkbook(Workbook):
row_value
[
next_col
]
=
'{0}
\n
{1}'
.
format
(
cn_str
,
row_value
[
next_col
])
return
row_value
@staticmethod
def
rm_second_row
(
row_value
,
amount_cell_idx
,
over_cell_idx
):
row_value
=
list
(
row_value
)
if
isinstance
(
over_cell_idx
,
int
)
and
isinstance
(
amount_cell_idx
,
int
):
max_idx
=
max
(
over_cell_idx
,
amount_cell_idx
)
elif
isinstance
(
over_cell_idx
,
int
):
max_idx
=
over_cell_idx
elif
isinstance
(
amount_cell_idx
,
int
):
max_idx
=
amount_cell_idx
else
:
max_idx
=
0
if
max_idx
>
1
:
append_list
=
[]
for
i
in
range
(
2
,
max_idx
+
1
):
if
isinstance
(
row_value
[
i
],
str
):
split_list
=
row_value
[
i
]
.
split
(
'
\n
'
)
row_value
[
i
]
=
split_list
[
0
]
append_list
.
extend
(
split_list
[
1
:])
if
isinstance
(
row_value
[
1
],
str
):
append_list
.
insert
(
0
,
row_value
[
1
])
row_value
[
1
]
=
'
\n
'
.
join
(
append_list
)
return
row_value
def
build_month_sheet
(
self
,
ms
,
card
,
month_mapping
,
is_reverse
,
statistics_header_info
,
max_column
,
classify
):
summary_cell_idx
=
statistics_header_info
.
get
(
consts
.
SUMMARY_KEY
)
date_cell_idx
=
statistics_header_info
.
get
(
consts
.
DATE_KEY
)
...
...
@@ -357,9 +380,11 @@ class BSWorkbook(Workbook):
for
row_value
in
ws
.
iter_rows
(
min_row
=
part
[
1
],
max_row
=
part
[
2
],
values_only
=
True
):
if
any
(
row_value
):
if
classify
==
consts
.
WECHART_CLASSIFY
:
row_value
=
self
.
wechart_func
(
row_value
,
*
consts
.
WECHART_ERROR_COL
)
row_value
=
self
.
rm_cn_char
(
row_value
,
*
consts
.
WECHART_ERROR_COL
)
elif
classify
==
consts
.
MS_CLASSIFY
:
row_value
=
self
.
wechart_func
(
row_value
,
*
consts
.
MS_ERROR_COL
)
row_value
=
self
.
rm_cn_char
(
row_value
,
*
consts
.
MS_ERROR_COL
)
elif
classify
==
consts
.
NYYH_CLASSIFY
:
row_value
=
self
.
rm_second_row
(
row_value
,
amount_cell_idx
,
over_cell_idx
)
new_ws
.
append
(
row_value
)
# 3.2.提取信息、高亮
amount_mapping
=
{}
...
...
Write
Preview
Styling with
Markdown
is supported
Attach a file
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to post a comment