merge wk branch concept of placing the mapping instructions in the regex items themselves

This commit is contained in:
Paul Trowbridge 2017-11-04 09:33:16 -04:00
parent 706fcd8c9c
commit 175b7a38c5
2 changed files with 420 additions and 180 deletions

View File

@ -9,25 +9,26 @@ SELECT
t.id,
t.rec,
m.target,
regex->>'map' map_intention,
m.seq,
regex->>'function' regex_function,
e.v ->> 'field' result_key_name,
e.v ->> 'key' target_json_path,
e.v ->> 'flag' regex_options_flag,
e.v->>'map' map_intention,
e.v->>'retain' retain_result,
e.v->>'regex' regex_expression,
e.rn target_item_number,
COALESCE(mt.rn,rp.rn,1) result_number,
mt.mt rx_match,
rp.rp rx_replace,
CASE regex->>'map'
WHEN 'yes' THEN
CASE e.v->>'map'
WHEN 'y' THEN
e.v->>'field'
ELSE
null
END map_key,
CASE regex->>'map'
WHEN 'yes' THEN
CASE e.v->>'map'
WHEN 'y' THEN
CASE regex->>'function'
WHEN 'extract' THEN
CASE WHEN array_upper(mt.mt,1)=1
@ -76,9 +77,9 @@ FROM
LEFT JOIN LATERAL regexp_replace(t.rec #>> ((e.v ->> 'key')::text[]), e.v ->> 'regex'::text, e.v ->> 'replace'::text,e.v ->> 'flag') WITH ORDINALITY rp(rp, rn) ON
m.regex->>'function' = 'replace'
WHERE
t.allj IS NULL
--t.srce = 'PNCC'
--rec @> '{"Transaction":"ACH Credits","Transaction":"ACH Debits"}'
t.map IS NULL
--rec @> '{"Description":"CHECK 93013270 086129935"}'::jsonb
ORDER BY
t.id DESC,
@ -87,13 +88,15 @@ ORDER BY
COALESCE(mt.rn,rp.rn,1)
)
--SELECT * FROM rx
--SELECT count(*) FROM rx LIMIT 100
, agg_to_target_items AS (
SELECT
srce
,id
,target
,seq
,map_intention
,regex_function
,target_item_number
@ -133,6 +136,7 @@ GROUP BY
srce
,id
,target
,seq
,map_intention
,regex_function
,target_item_number
@ -142,13 +146,15 @@ GROUP BY
,retain_key
)
--SELECT * FROM agg_to_target_items
--SELECT * FROM agg_to_target_items LIMIT 100
, agg_to_target AS (
SELECT
srce
,id
,target
,seq
,map_intention
,tps.jsonb_concat_obj(COALESCE(map_val,'{}'::JSONB)) map_val
,jsonb_strip_nulls(tps.jsonb_concat_obj(COALESCE(retain_val,'{}'::JSONB))) retain_val
@ -158,6 +164,7 @@ GROUP BY
srce
,id
,target
,seq
,map_intention
ORDER BY
id
@ -172,6 +179,7 @@ SELECT
a.srce
,a.id
,a.target
,a.seq
,a.map_intention
,a.map_val
,a.retain_val retain_value
@ -190,7 +198,7 @@ FROM
SELECT
srce
,id
,tps.jsonb_concat_obj(COALESCE(retain_value,'{}'::jsonb)) retain_val
,tps.jsonb_concat_obj(COALESCE(retain_value,'{}'::jsonb) ORDER BY seq DESC) retain_val
,tps.jsonb_concat_obj(COALESCE(map,'{}'::jsonb)) map
FROM
link_map
@ -199,7 +207,8 @@ GROUP BY
,id
)
SELECT srce, id, jsonb_pretty(retain_val), jsonb_pretty(map) FROM agg_to_id
--SELECT agg_to_id.srce, agg_to_id.id, jsonb_pretty(agg_to_id.retain_val) , jsonb_pretty(agg_to_id.map) FROM agg_to_id ORDER BY id desc LIMIT 100
UPDATE

View File

@ -1,11 +1,6 @@
/*
DELETE FROM tps.map_rm where target = 'Strip Amount Commas';
INSERT INTO
tps.map_rm
SELECT *
FROM
(VALUES
('PNCC', 'Strip Amount Commas',
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"Strip Amount Commas",
@ -17,263 +12,499 @@ FROM
"regex": ",",
"replace":"",
"flag":"g",
"retain":"y"
"retain":"y",
"map":"n"
}
],
"function":"replace",
"map":"no",
"where": [
{
}
]
}
$j$::jsonb
, 1)
) x;
WHERE
target = 'Strip Amount Commas';
DELETE FROM tps.map_rm where target = 'Parse Descr';
INSERT INTO
tps.map_rm
SELECT *
FROM
(VALUES
('PNCC', 'Parse Descr',
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"Parse Descr",
"description":"parse the description based on at least three capital letters followed by a comma until another set of at lesat 3 capital letters and a comma is encountered",
"defn": [
{
"key": "{Description}",
"field": "dparse",
"regex": "([A-Z]{3,}?:)(.*)(?=[A-Z]{3,}?:|$)",
"flag":"g",
"retain":"y"
}
],
"function":"extract",
"map":"no",
"where": [
{
}
]
}
$j$::jsonb
, 2)
) x;
DELETE FROM tps.map_rm where target = 'Extract OBI';
INSERT INTO
tps.map_rm
SELECT *
FROM
(VALUES
('PNCC', 'Extract OBI',
$j$
{
"name":"Extract OBI",
"description":"pull out whatever follows OBI in the description until atleast 3 capital letters followed by a colon are encountered",
"defn": [
{
"key": "{Description}",
"field": "obi",
"regex": "OBI:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y"
}
],
"function":"extract",
"map":"no",
"where": [
{
"Transaction":"Money Transfer DB - Wire"
},
{
"Transaction":"Money Transfer CR-Other"
},
{
"Transaction":"Intl Money Transfer Debits"
},
{
"Transaction":"Money Transfer DB - Other"
},
{
"Transaction":"Money Transfer CR-Wire"
}
]
}
$j$::jsonb
, 2)
) x;
DELETE FROM tps.map_rm where target = 'Extract RFB';
INSERT INTO
tps.map_rm
SELECT *
FROM
(VALUES
('PNCC', 'Extract RFB',
$j$
{
"name":"Extract RFB",
"description":"pull out whatever follows RFB in the description until atleast 3 capital letters followed by a colon are encountered",
"defn": [
{
"key": "{Description}",
"field": "rfb",
"regex": "RFB:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y"
}
],
"function":"extract",
"map":"no",
"where": [
{
"Transaction":"Money Transfer DB - Wire"
},
{
"Transaction":"Money Transfer CR-Other"
},
{
"Transaction":"Intl Money Transfer Debits"
},
{
"Transaction":"Money Transfer DB - Other"
},
{
"Transaction":"Money Transfer CR-Wire"
}
]
}
$j$::jsonb
, 2)
) x;
DELETE FROM tps.map_rm where target = 'Parse ACH';
INSERT INTO
tps.map_rm
SELECT *
FROM
(VALUES
('PNCC', 'Parse ACH',
$j$
{
"name":"Parse ACH",
"name":"Parse ACH Credits",
"description":"parse select components of the description for ACH Credits Receieved",
"defn": [
{
"key": "{Description}",
"field":"Comp Name",
"regex": "Comp Name:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"field":"beneficiary",
"regex": "Comp Name:(.+?)(?=\\d{6} Com|SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Cust ID",
"regex": "Cust ID:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Desc",
"regex": "Desc:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Cust Name",
"field":"originator",
"regex": "Cust Name:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Batch Discr",
"regex": "Batch Discr:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Comp ID",
"regex": "Comp ID:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Addenda",
"regex": "Addenda:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"SETT",
"regex": "SETT:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Date",
"regex": "Date:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Time",
"regex": "Time:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
}
],
"function":"extract",
"map":"no",
"where": [
{
"Transaction":"ACH Credits"
}
]
}
$j$::jsonb
WHERE target = 'Parse ACH Credits';
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"Parse ACH Debits",
"description":"parse select components of the description for ACH Credits Receieved",
"defn": [
{
"key": "{Description}",
"field":"originator",
"regex": "Comp Name:(.+?)(?=\\d{6} Com|SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Cust ID",
"regex": "Cust ID:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Desc",
"regex": "Desc:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"beneficiary",
"regex": "Cust Name:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Batch Discr",
"regex": "Batch Discr:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Comp ID",
"regex": "Comp ID:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Addenda",
"regex": "Addenda:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"SETT",
"regex": "SETT:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Date",
"regex": "Date:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field":"Time",
"regex": "Time:(.+?)(?=SEC:|Cust ID:|Desc:|Comp Name:|Comp ID:|Batch Discr:|Cust Name:|Addenda:|SETT:|Date:|Time:|$)",
"flag":"",
"retain":"y",
"map":"n"
}
],
"function":"extract",
"where": [
{
"Transaction":"ACH Debits"
}
]
}
$j$::jsonb
, 2)
) x;
*/
WHERE target = 'Parse ACH Debits';
UPDATE
tps.map_rm
SET
regex =
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"First 20",
"description":"pull first 20 characters from description for mapping",
"name":"Parse Wires",
"description":"pull out whatever follows OBI in the description until atleast 3 capital letters followed by a colon are encountered",
"defn": [
{
"key": "{Memo}",
"field":"f20",
"regex": ".{1,20}",
"key": "{Description}",
"field": "dparse",
"regex": "([A-Z]{3,}?:)(.*)(?=[A-Z]{3,}?:|$)",
"flag":"g",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "beneficiary_components",
"regex": "BENEFICIARY:(.*?)AC/(\\d*) (.*)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y"
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "originator_components",
"regex": "ORIGINATOR:(.*?)AC/(\\d*) (.*)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "beneficiary",
"regex": "BENEFICIARY:(.*?)AC/\\d* .*(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "originator",
"regex": "ORIGINATOR:(.*?)AC/\\d* .*(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "OBI",
"regex": "OBI:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "RFB",
"regex": "RFB:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "ABA",
"regex": "ABA:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "BBI",
"regex": "BBI:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "BENEBNK",
"regex": "BENEBNK:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "IBK",
"regex": "IBK:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "RATE",
"regex": "RATE:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
},
{
"key": "{Description}",
"field": "RECVBNK",
"regex": "RECVBNK:(.*?)(?=[A-Z]{3,}?:|$)",
"flag":"",
"retain":"y",
"map":"n"
}
],
"function":"extract",
"where": [
{
"Transaction":"Money Transfer DB - Wire"
},
{
"Transaction":"Money Transfer DB - Other"
},
{
"Transaction":"Money Transfer CR-Wire"
},
{
"Transaction":"Money Transfer CR-Other"
},
{
"Transaction":"Intl Money Transfer Debits"
},
{
"Transaction":"Intl Money Transfer Credits"
}
]
}
$j$::jsonb
WHERE target = 'Parse Wires';
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"Trans Type",
"description":"extract intial description in conjunction with account name and transaction type for mapping",
"defn": [
{
"key": "{AccountName}",
"field": "acctn",
"regex": "(.*)",
"retain": "n",
"map":"y"
},
{
"key": "{Transaction}",
"field": "trans",
"regex": "(.*)",
"retain": "n",
"map":"y"
},
{
"key": "{Description}",
"field": "ini",
"regex": "([\\w].*?)(?=$| -|\\s[0-9].*?|\\s[\\w/]+?:)",
"retain": "y",
"map":"y"
}
],
"where": [
{
}
],
"function": "extract"
}
$j$::jsonb
WHERE target = 'Trans Type';
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"Currency",
"description":"pull out currency indicators from description of misc items and map",
"defn": [
{
"key": "{Description}",
"field": "ini",
"regex": "([\\w].*?)(?=$| -|\\s[0-9].*?|\\s[\\w/]+?:)",
"retain": "y",
"map":"y"
},
{
"key": "{Description}",
"field": "curr1",
"regex": ".*(DEBIT|CREDIT).*(USD|CAD).*(?=DEBIT|CREDIT).*(?=USD|CAD).*",
"retain": "y",
"map":"y"
},
{
"key": "{Description}",
"field": "curr2",
"regex": ".*(?=DEBIT|CREDIT).*(?=USD|CAD).*(DEBIT|CREDIT).*(USD|CAD).*",
"retain": "y",
"map":"y"
}
],
"where": [
{
"Transaction": "Miscellaneous Credits"
},
{
"Transaction": "Miscellaneous Debits"
}
],
"function": "extract"
}
$j$::jsonb
WHERE target = 'Currency';
UPDATE tps.map_rm
SET regex =
$j$
{
"defn": [
{
"key": "{Description}",
"field": "checkn",
"regex": "[^0-9]*([0-9]*)\\s|$",
"retain": "y",
"map":"n"
}
],
"where": [
{
"Transaction": "Checks Paid"
}
],
"function": "extract"
}
$j$::jsonb
WHERE target = 'Check Number';
UPDATE tps.map_rm
SET regex =
$j$
{
"name":"ADP Codes",
"description":"link to adp code definitions",
"defn": [
{
"key": "{gl_descr}",
"field": "gl_descr",
"regex": ".*",
"flag":"",
"retain":"n",
"map":"y"
},
{
"key": "{prim_offset}",
"field": "prim_offset",
"regex": ".*",
"flag":"",
"retain":"n",
"map":"y"
},
{
"key": "{pay_date}",
"field": "pay_month",
"regex": ".{1,4}",
"flag":"",
"retain":"y",
"map":"n"
}
],
"function":"extract",
"map":"yes",
"where": [
{
}
]
}
$j$::jsonb
WHERE srce = 'HUNT'
WHERE
target = 'ADP Codes';