Test Autofill's ability to merge duplicate profiles and

throw away junk profiles. Includes a dataset converter script to convert csv file into profile dictionary list. testMergeDuplicateProfilesInAutofill Added additional tests: testFilterMalformedEmailAddresses - covers fixed bug 73654. testFilterIncompleteAddresses - covers fixed bug 71710. BUG=none TEST=none Review URL: http://codereview.chromium.org/6246147 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@75344 0039d316-1c4b-4281-b951-d872f2087c98

Test Autofill's ability to merge duplicate profiles and
throw away junk profiles. Includes a dataset converter script to convert csv file into profile dictionary list. testMergeDuplicateProfilesInAutofill Added additional tests: testFilterMalformedEmailAddresses - covers fixed bug 73654. testFilterIncompleteAddresses - covers fixed bug 71710. BUG=none TEST=none Review URL: http://codereview.chromium.org/6246147 git-svn-id: svn://svn.chromium.org/chrome/trunk/src@75344 0039d316-1c4b-4281-b951-d872f2087c98
a57325e1 · dyu@chromium.org · b964574e · a57325e1 · a57325e1 · a57325e1
Commit a57325e1 authored Feb 18, 2011 by dyu@chromium.org
5 changed files
--- a/chrome/test/data/autofill/dataset.txt
+++ b/chrome/test/data/autofill/dataset.txt
+This dataset file can be parsed with dataset_converter.py
+
+first_name, middle_name, last_name, email, company_name, address_line_1, address_line_2, city, state, zipcode, country, phone, fax
+
+John||Doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|California|14888|US|4195551234|
+John||Doe|john.doe@movitron.com|Movitron|Iceberg @ 10 Pennsylvania|Movitron 5th Floor|Mountain View|CA|14890|US|4195551234|
+John||Doe|john@incarnate.net||||SFO|-08:00 Pacific Time (US & Canada); Tijuana||US||
+Aman||Radmon||Current or past|||Miles|North America||United States||
+|||||16th btw Mission & Valencia||||14888||4195551234|
+John|K|Doe|John.doe@gmail.com||||||Postal Code|||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave 2|||||AF||
+JOHN||DOE||||||||||
+Movitron|||||10 Pennsylvania St.|at Spear|Mountain View||14889|CA||
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+John||Doe|john.doe@gmail.com|self|Movitron 5th Floor||Mountain View|CA|14891|US|4195551234|
+Annie||Doe||||||||||
+John||Doe|john.doe@gmail.com|Movitron|1950 Amphitheatre #2||Mountain View|CA|14888|United States|4195551234|
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+John M Doe|US|false|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|john.doe@gmail.com|john.doe@gmail.com|Mountain View|CA|14888|US||
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|||CA||USA|4195551234|
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com||CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+John|K|Doe|john.doe@movitron.com|||||||||
+James||Ralph|john.doe@gmail.com|CP|||||14888|us||
+John||Doe|1|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com|||14888||4195551234|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|john.doe@gmail.com|john.doe@gmail.com|Mountain View|CA|14888|US||
+John|K|Doe|John.doe@gmail.com||||||Postal Code|||
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+John||Doe|john@incarnate.net|C|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John||Doe|john.doe@gmail.com||john.doe@gmail.com|john.doe@gmail.com|||14888|81||
+John||Doe|john.doe@gmail.com|Movitron|1950 Amphitheatre #2||California|Commercial/Industrial|14888|81|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|16th btw Mission & Valencia||Mountain View|CA|14888|US|4195551234|
+cardinal||||CP|||||17505|us||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave #2||||14888|AF||
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+||||Choose a company...||||||-||
+John|K|Doe|john.doe@gmail.com||||||Postal Code|United States|4195551234|
+Doe||House|||827 Peaches||Ben|Tx|53077|||
+DodoLand|||||827 Peaches||Ben|Tx|53077|||
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com||CA|14888|USA|4195551234|
+John|K|Doe|john.doe@gmail.com||||||Postal Code||4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+John||Doe|john.doe@gmail.com|self|Movitron 5th Floor||Mountain View|CA|14891|US|4195551234|
+John||Doe|john.doe+goto@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+James||Ralph|john.doe@gmail.com|CP|||||14888|us||
+||||lowercase capital||||||dk||
+||||lowercase capital||||||dk||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave #2||||14888|AF||
+JOHN||DOE||||||||||
+Annie||Doe||||||||||
+John doe|1950 Amphitheatre #2|Doe|john.doe@gmail.com||||Mountain View|CA|14888|US|5551234|
+John doe|1950 Amphitheatre #2|Doe|john.doe@gmail.com||||Mountain View|CA|14888|US|5551234|
+John||Doe|john.doe+goto@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+John||Doe|john@incarnate.net|CP|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John|K|Doe|john.doe@movitron.com|||||||||
+john|k|doe|john.doe@gmail.com|||true|Mountain View|9110|14891|4601|4195551234|
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave 2|||||AF||
+John||Doe|PILOT PROGRAM - wireless||1950 Amphitheatre Ave #2||7x54lbbs||14888|us|4195551234|
+||||lowercase capital||||||dk||
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John M Doe|US|false|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||Mountian View|9110|14888|4601|4195551234|
+|||||9128 s. Broadway|||TX|53077|||
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||||14888|us|4195551234|
+john||Doe|john.doe@gmail.com||||||14888|||
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|United States||
+John||Doe|john.doe@gmail.com|www.movitron.com||||||US|4195551234|
+|||||9128 s. Broadway|||TX|53077|||
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||Mountian View|9110|14888|4601|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+|||||9128 s. Broadway|||TX|53077|||
+John M Doe|US|false|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||||14888|us|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John||Doe|john.doe@gmail.com|||||||1||
+john||Doe|john.doe@gmail.com||||||14888|||
+|||||||MOUNTAINVIEW|California||United States||
+||||Movitron, Inc.|Iceberg @ 10 Pennsylvania|Movitron 5th Fl|Mountain View|California|14891|United States|4195551234|
+john||Doe|john.doe@gmail.com|myself only||||CA|14888|myself only||
+John||Doe|john.doe@gmail.com||||||14888|81||
+John||Doe|Evening||1950 Amphitheatre Ave #2||Mountain View||14888|Belmont|4195551234|
+John||Doe|Every New Message|Movitron, Inc.|Iceberg @ 10 Pennsylvania||Mountain View|CA|14888|US|4195551234|
+John||Doe|john.doe@gmail.com|Movitron|1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||doe|john.doe@gmail.com||||||14888|14888||
+cardinal||||CP|||||17505|us||
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|john.doe@gmail.com|john.doe@gmail.com|Mountain View|CA|14888|US||
+John|K|Doe|John.doe@gmail.com||||||Postal Code|||
+John|K|Doe|john.doe@movitron.com|||||||||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave 2|||||AF||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave #2||||14888|AF||
+JOHN||DOE||||||||||
+John||Doe|john.doe@gmail.com|self|btw bryant and forida|||CA||US|4195551234|
+James||Ralph|john.doe@gmail.com|CP|||||14888|us||
+James||Ralph|john.doe@gmail.com|CP|||||14888|us||
+John||Doe|john.doe@gmail.com|self|16th & mission|||CA||US|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+Annie||Doe||||||||||
+John doe|1950 Amphitheatre #2|Doe|john.doe@gmail.com||||Mountain View|CA|14888|US|5551234|
+John||Doe|john.doe+goto@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+James||Ralph|john.doe@gmail.com|CP|||||14888|us||
+||||lowercase capital||||||dk||
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com||CA|14888|USA|4195551234|
+John|K|Doe|john.doe@gmail.com||||||Postal Code||4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+John||Doe|john.doe@gmail.com|self|Movitron 5th Floor||Mountain View|CA|14891|US|4195551234|
+AT&T||Mobility|||AT&T Mobility|PO BOX 897234|Santa Cruz|CA|12395||8005557612|
+John M Doe|US|false|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+john|k|doe|john.doe@gmail.com|||true|Mountain View|9110|14891|4601|4195551234|
+John M Doe|US|false|john.doe@gmail.com||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+John doe|req|the recipient's shipping Name|||||Mountain View|req|14888|CA|4195551234|
+John||Doe|john.doe@gmail.com||john.doe@gmail.com|john.doe@gmail.com|||14888|81||
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+AT&T||Mobility|||AT&T Mobility|PO BOX 897234|Santa Cruz||12345||8005557612|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||||14888|us|4195551234|
+John|K|Doe|john.doe@gmail.com||||Mountian View|9110|14888|4601|4195551234|
+John|K|Doe|john.doe@gmail.com||||||14888|us|4195551234|
+John|K|Doe|john.doe@gmail.com||16th btw Mission & Valencia||Mountian View|9110|14888|4601|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+cardinal|||john.doe@gmail.com|CP|||||14888|us||
+John||Doe|1|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com|||14888||4195551234|
+john|k|doe|john.doe@gmail.com|||true|Mountain View|9110|14891|4601|4195551234|
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+cardinal||||CP|||||17505|us||
+James||Ralph|john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+john||doe|john.doe@gmail.com||1950 Amphitheatre Ave #2||||14888|US|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+1|||john.doe@gmail.com|||||||2||
+John||Doe|john.doe@gmail.com|self|btw bryant and forida|||CA|14888|US|4195551234|
+John||Doe|john.doe@gmail.com|self|16th & mission|||CA|14888|US|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+1|||john.doe@gmail.com||||||14888|2||
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+john||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+|||html|||||||||
+john||Doe|john.doe@gmail.com||||||14888|||
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|16th btw Mission & Valencia||Mountain View|CA|14888|US|4195551234|
+John doe|req|the recipient's shipping Name|john.doe@gmail.com||16th btw Mission & Valencia||Mountain View|req|14888|CA|4195551234|
+john||doe|john.doe@gmail.com||16th btw Mission & Valencia||||14888|14888|4195551234|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|16th btw Mission & Valencia||Mountain View|CA|14888|US|4195551234|
+cardinal|||john.doe@gmail.com|CP|16th btw Mission & Valencia||||14888|us|4195551234|
+John||Doe|gmail||16th btw Mission & Valencia||||14888|US|4195551234|
+john|k|doe|john.doe@gmail.com||16th btw Mission & Valencia|true|||14888|us|4195551234|
+John||Doe|1|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com|||14888||4195551234|
+John||Doe|john.doe@gmail.com||john.doe@gmail.com|john.doe@gmail.com|||14888|81||
+John||Doe|1|Movitron, Inc.|john.doe@gmail.com||||||4195551234|
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|||CA||USA|4195551234|
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|||CA||USA|4195551234|
+John||Doe|john.doe@movitron.com|Movitron, Inc.|john.doe@gmail.com|||||Mountain View, CA|4195551234|
+||||lowercase capital||||||dk||
+John||Doe|john.doe@movitron.com|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com|||14888|Mountain View, CA|4195551234|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|john.doe@gmail.com|john.doe@gmail.com|Mountain View|CA|14888|US||
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com||CA|14888|USA|4195551234|
+John||Doe|gmail||john.doe@gmail.com|john.doe@gmail.com|||14888|US|4195551234|
+John||Doe|john.doe@gmail.com|Movitron, Inc. / Compa.ny|john.doe@gmail.com|john.doe@gmail.com|Mountain View|CA|14888|US||
+John||Doe|4195551234|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com||CA|14888|USA|4195551234|
+John||Doe|john.doe@movitron.com|Movitron, Inc.|example.com||||||4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+||||lowercase capital||||||dk||
+||||lowercase capital||||||dk||
+John||Doe|john.doe@gmail.com|self|16th & mission||7x54lbbs|CA|14888|US|4195551234|
+John||Doe|john.doe@gmail.com|self|btw bryant and forida||7x54lbbs|CA|14888|US|4195551234|
+John||Doe|john.doe+goto@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+John||Doe|PILOT PROGRAM - wireless||1950 Amphitheatre Ave #2||7x54lbbs||14888|us|4195551234|
+John||Doe|john.doe@gmail.com|self|btw Mission and Valencia off 16th||7x54lbbs|CA|14888|US|4195551234|
+John||Doe|john.doe@gmail.com|self|Movitron 5th Floor||Mountain View|CA|14891|US|4195551234|
+John||Doe|419.555.1234||1950 Amphitheatre #2|||CA|14888|US||
+John|K|Doe|john.doe@gmail.com||||||Postal Code||4195551234|
+John doe|1950 Amphitheatre #2|Doe|john.doe@gmail.com||||Mountain View|CA|14888|US|5551234|
+John||Doe|gmail|Movitron, Inc.|john.doe@gmail.com|john.doe@gmail.com|7x54lbbs||14888|US|4195551234|
+Annie||Doe||||||||||
+JOHN||DOE||||||||||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave #2||||14888|AF||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave 2|||||AF||
+John|K|Doe|john.doe@movitron.com|||||||||
+John||Doe|john@incarnate.net|CP|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John||Dodo|john.dodo@gmail.com||||Mountain View||14888||4195551236|
+John||Doe|john.doe+goto@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+John doe|1950 Amphitheatre #2||john.doe@gmail.com||||||14888|US||
+John doe|1950 Amphitheatre #2|Doe|john.doe@gmail.com||||Mountain View|CA|14888|US|5551234|
+John||Doe|PILOT PROGRAM - wireless||1950 Amphitheatre Ave #2||||||4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave #2||||14888|AF||
+John|K|Doe|John.doe@gmail.com||||||Postal Code|||
+John|K|Doe|john.doe@movitron.com|||||||||
+John||Doe|john.doe@gmail.com|self|btw Mission and Valencia off 16th|||CA|14888|US|4195551234|
+John|K|Doe|john.doe@gmail.com||||||Postal Code||4195551234|
+John||Doe|john@incarnate.net|C|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John||Doe|john.doe@gmail.com|self|Movitron 5th Floor||Mountain View|CA|14891|US|4195551234|
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+John||Doe|PILOT PROGRAM - wireless||1950 Amphitheatre Ave #2||Mountain View|9110|14891|4601|4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|CA|14888|USA|4195551234|
+John||Doe|john@incarnate.net|CP|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John||Doe|PILOT PROGRAM - wireless||1950 Amphitheatre Ave #2||7x54lbbs||14888|us|4195551234|
+John|K|Doe|John.doe@gmail.com||||||Postal Code|||
+John M Doe|US|false|||1950 Amphitheatre Ave #2||Mountain View|CA|14888|US|2935190284|
+John doe|1950 Amphitheatre #2||john.doe@gmail.com|Movitron, Inc.|Iceberg @ 10 Pennsylvania 5th Floor||Mountain View|CA|14891|US|4195551234|
+JOHN||DOE||||||||||
+Betty||Blue|betty.blue@gmail.com||1950 Amphitheatre Ave 2|||||AF||
+Annie||Doe||||||||||
+John|K|Doe|john.doe@movitron.com||1950 Amphitheatre Ave|No. 2|Mountain View|CA|14891|||
+John||Doe|john@incarnate.net|C|1950 Amphitheatre #2||Mountain View|CA|14888|us|4195551234|
+John Doe|john.doe@gmail.com|web resume||||||||||
+John||Doe|419.555.1234||1950 Amphitheatre #2||7x54lbbs|CA|14888|US||
+John||Doe|john.doe@movitron.com|26 or more|1950 Amphitheatre #2||Mountain View|CA|14888|US|4195551234|
+JOHN|K|DOE|john.doe@gmail.com||10 Pennsylvania 2nd floor||Mountain View|CA|14777|US|4195551234|
+JOHN|–|DOE|john.doe@movitron.com||10 Pennsylvania 2nd floor|Market & Spear|Mountain View|CA|14777|US|4195551234|
+Movitron|||||10 Pennsylvania St.|at Spear st|Mountain View||14889|CA||
+Movitron|||||10 Pennsylvania St.|at Spear|Mountain View||14889|CA||
+JOHN|K|DOE|john.doe@gmail.com||10 Pennsylvania 2nd floor||Mountain View|CA|14891|US|4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre #2||Mountain View|California|14888|United States|4195551234|
+John|Doe|john.doe@gmail.com||||||||||
+John||Doe|john.doe@gmail.com|Movitron, Inc.|1950 Amphitheatre #2||Mountain View|5551768|14888|27701|4195551234|
+John||Doe|john.doe@gmail.com||1950 Amphitheatre Ave Apt 2||Mountain View|CA|14888|Mountain View, CA|4195551234|
+subdomain.dreamhost.com|||john.doe@gmail.com||1950 Amphitheatre #2||Mountin View|CA|14890|US|4195551234|
+Movitron|– Champlain Meeting|Room|||10 Pennsylvania 5th floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Pythagoras Telepresence|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Champlain Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Euclid Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Copernicus Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Newton Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Ockham Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Archimedes Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+Movitron|– Brahmagupta Meeting|Room|||10 Pennsylvania 2nd floor|Market & Spear|Mountain View||14891|USA||
+John||Doe|john.doe@gmail.com|Movitron, Inc.|Iceberg @ 10 Pennsylvania 5th Floor||Mountain View|CA|14891|USA|4195551234|
+|||||||Miles|WORLD||||
--- a/chrome/test/data/autofill/duplicate_profiles_test.html
+++ b/chrome/test/data/autofill/duplicate_profiles_test.html
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN">
+<html>
+  <head>
+    <title>AutoFill Form</title>
+  </head>
+  <body>
+    <form id="merge_dup" method="post">
+      <p>
+      <!-- Profile -->
+      <!-- The form element names must match the keys in
+      dataset_duplicate-profiles.txt -->
+        <label for="NAME_FIRST">First Name:</label> 
+        <input type="text" id="NAME_FIRST" name="firstname"><br/>
+        <label for="NAME_MIDDLE">Middle Name:</label> 
+        <input type="text" id="NAME_MIDDLE" name="middlename"><br/>
+        <label for="NAME_LAST">Last Name:</label> 
+        <input type="text" id="NAME_LAST" name="lastname"><br/>
+        <label for="EMAIL_ADDRESS">Email:</label> 
+        <input type="text" id="EMAIL_ADDRESS" name="email"><br/>
+        <label for="COMPANY_NAME">Company:</label> 
+        <input type="text" id="COMPANY_NAME" name="company"><br/>
+
+        <label for="ADDRESS_HOME_LINE1">Address:</label> 
+        <input type="text" id="ADDRESS_HOME_LINE1" name="address"><br/>
+        <label for="ADDRESS_HOME_LINE2">Address 2:</label> 
+        <input type="text" id="ADDRESS_HOME_LINE2" name="address2"><br/>
+        <label for="ADDRESS_HOME_CITY">City:</label> 
+        <input type="text" id="ADDRESS_HOME_CITY" name="city"><br/>
+        <label for="ADDRESS_HOME_STATE">State:</label> 
+        <input type="text" id="ADDRESS_HOME_STATE" name="state"><br/>
+        <label for="ADDRESS_HOME_ZIP">Zip:</label> 
+        <input type="text" id="ADDRESS_HOME_ZIP" name="zipcode"><br/>
+
+        <label for="ADDRESS_HOME_COUNTRY">Country:</label> 
+        <input type="text" id="ADDRESS_HOME_COUNTRY" name="country"><br/>
+        <label for="PHONE_HOME_WHOLE_NUMBER">Phone:</label> 
+        <input type="text" id="PHONE_HOME_WHOLE_NUMBER" name="phone"><br/>
+        <label for="PHONE_FAX_WHOLE_NUMBER">Fax:</label> 
+        <input type="text" id="PHONE_FAX_WHOLE_NUMBER" name="fax"><br/>
+        <input type="submit" value="send"> <input type="reset">
+      </p>
+    </form>
+  </body>
+</html>
--- a/chrome/test/functional/PYAUTO_TESTS
+++ b/chrome/test/functional/PYAUTO_TESTS
@@ -29,7 +29,10 @@
      'autofill',
      # Not part of the continous build.
      # Used for testing Autofill crowdsourcing.
-      '-autofill.AutoFillTest.testAutofillCrowdSourcing',
+      '-autofill.AutoFillTest.testAutofillCrowdsourcing',
+      # Not part of the continous build.
+      # Used for testing Autofill duplicate profile merging.
+      '-autofill.AutoFillTest.testMergeDuplicateProfilesInAutofill',
      'bookmark_bar',
      'bookmarks',
      'browser',

--- a/chrome/test/functional/autofill.py
+++ b/chrome/test/functional/autofill.py
@@ -3,8 +3,11 @@
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.

+import logging
 import os
+import pickle

+import dataset_converter
 import pyauto_functional  # Must be imported before pyauto
 import pyauto

@@ -93,10 +96,63 @@ class AutoFillTest(pyauto.PyUITest):
    self.assertEqual([expected_credit_card],
                     self.GetAutoFillProfile()['credit_cards'])

-  def testAutofillCrowdSourcing(self):
-    """Test able to send POST request of web form to crowd source server.
-    Require a loop of 1000 submits as the source server only collects 1% of
-    the data posted."""
+  def testFilterIncompleteAddresses(self):
+    """Test Autofill filters out profile with incomplete address info."""
+    profile = {'NAME_FIRST': 'Bob',
+               'NAME_LAST': 'Smith',
+               'EMAIL_ADDRESS': 'bsmith@example.com',
+               'COMPANY_NAME': 'Company X',
+               'PHONE_HOME_WHOLE_NUMBER': '650-123-4567',}
+    url = self.GetHttpURLForDataPath(
+        os.path.join('autofill', 'dup-profiles-test.html'))
+    self.NavigateToURL(url)
+    for key, value in profile.iteritems():
+      script = ('document.getElementById("%s").value = "%s"; '
+                'window.domAutomationController.send("done");') % (key, value)
+      self.ExecuteJavascript(script, 0, 0)
+    js_code = """
+      document.getElementById("merge_dup").submit();
+      window.addEventListener("unload", function() {
+        window.domAutomationController.send("done");
+      });
+    """
+    self.ExecuteJavascript(js_code, 0, 0)
+    self.assertEqual([], self.GetAutoFillProfile()['profiles'])
+
+  def testFilterMalformedEmailAddresses(self):
+    """Test Autofill filters out malformed email address during form submit."""
+    profile = {'NAME_FIRST': 'Bob',
+               'NAME_LAST': 'Smith',
+               'EMAIL_ADDRESS': 'garbage',
+               'ADDRESS_HOME_LINE1': '1234 H St.',
+               'ADDRESS_HOME_CITY': 'San Jose',
+               'ADDRESS_HOME_STATE': 'CA',
+               'ADDRESS_HOME_ZIP': '95110',
+               'COMPANY_NAME': 'Company X',
+               'PHONE_HOME_WHOLE_NUMBER': '408-123-4567',}
+    url = self.GetHttpURLForDataPath(
+        os.path.join('autofill', 'dup-profiles-test.html'))
+    self.NavigateToURL(url)
+    for key, value in profile.iteritems():
+      script = ('document.getElementById("%s").value = "%s"; '
+                'window.domAutomationController.send("done");') % (key, value)
+      self.ExecuteJavascript(script, 0, 0)
+    js_code = """
+      document.getElementById("merge_dup").submit();
+      window.addEventListener("unload", function() {
+        window.domAutomationController.send("done");
+      });
+    """
+    self.ExecuteJavascript(js_code, 0, 0)
+    if 'EMAIL_ADDRESS' in self.GetAutoFillProfile()['profiles'][0]:
+      raise KeyError('TEST FAIL: Malformed email address is saved in profiles.')
+
+  def testAutofillCrowdsourcing(self):
+    """Test able to send POST request of web form to Autofill server.
+
+    The Autofill server processes the data offline, so it can take a few days
+    for the result to be detectable. Manual verification is required.
+    """
    # HTML file needs to be run from a specific http:// url to be able to verify
    # the results a few days later by visiting the same url.
    url = 'http://www.corp.google.com/~dyu/autofill/crowdsourcing-test.html'
@@ -105,24 +161,60 @@ class AutoFillTest(pyauto.PyUITest):
                             'crowdsource_autofill.txt')
    profiles = self.EvalDataFrom(file_path)
    self.FillAutoFillProfile(profiles=profiles)
+    # Autofill server captures 2.5% of the data posted.
+    # Looping 1000 times is a safe minimum to exceed the server's threshold or
+    # noise.
    for i in range(1000):
      fname = self.GetAutoFillProfile()['profiles'][0]['NAME_FIRST']
      lname = self.GetAutoFillProfile()['profiles'][0]['NAME_LAST']
      email = self.GetAutoFillProfile()['profiles'][0]['EMAIL_ADDRESS']
      # Submit form to collect crowdsourcing data for Autofill.
      self.NavigateToURL(url, 0, 0)
-      fname_field = 'document.getElementById("fn").value = "%s"; ' \
-                    'window.domAutomationController.send("done")' % fname
-      lname_field = 'document.getElementById("ln").value = "%s"; ' \
-                    'window.domAutomationController.send("done")' % lname
-      email_field = 'document.getElementById("em").value = "%s"; ' \
-                    'window.domAutomationController.send("done")' % email
+      fname_field = ('document.getElementById("fn").value = "%s"; '
+                     'window.domAutomationController.send("done");') % fname
+      lname_field = ('document.getElementById("ln").value = "%s"; '
+                     'window.domAutomationController.send("done");') % lname
+      email_field = ('document.getElementById("em").value = "%s"; '
+                     'window.domAutomationController.send("done");') % email
      self.ExecuteJavascript(fname_field, 0, 0);
      self.ExecuteJavascript(lname_field, 0, 0);
      self.ExecuteJavascript(email_field, 0, 0);
      self.ExecuteJavascript('document.getElementById("frmsubmit").submit();'
-                             'window.domAutomationController.send("done")',
+                             'window.domAutomationController.send("done");',
+                             0, 0)
+
+  def testMergeDuplicateProfilesInAutofill(self):
+    """Test Autofill ability to merge duplicate profiles and throw away junk."""
+    # HTML file needs to be run from a http:// url.
+    url = self.GetHttpURLForDataPath(
+        os.path.join('autofill', 'duplicate_profiles_test.html'))
+    # Run the parser script to generate the dictionary list needed for the
+    # profiles.
+    c = dataset_converter.DatasetConverter(
+        os.path.join(self.DataDir(), 'autofill', 'dataset.txt'),
+        logging_level=logging.INFO)  # Set verbosity to INFO, WARNING, ERROR.
+    list_of_dict = c.Convert()
+
+    for profile in list_of_dict:
+      self.NavigateToURL(url)
+      for key, value in profile.iteritems():
+        script = ('document.getElementById("%s").value = "%s"; '
+                  'window.domAutomationController.send("done");') % (key, value)
+        self.ExecuteJavascript(script, 0, 0)
+      self.ExecuteJavascript('document.getElementById("merge_dup").submit();'
+                             'window.domAutomationController.send("done");',
                             0, 0)
+    # Verify total number of inputted profiles is greater than the final number
+    # of profiles after merging.
+    self.assertTrue(
+        len(list_of_dict) > len(self.GetAutoFillProfile()['profiles']))
+    # Write profile dictionary to a file.
+    merged_profile = os.path.join(self.DataDir(), 'autofill',
+                                  'merged-profiles.txt')
+    profile_dict = self.GetAutoFillProfile()['profiles']
+    output = open(merged_profile, 'wb')
+    pickle.dump(profile_dict, output)
+    output.close()


 if __name__ == '__main__':

--- a/chrome/test/functional/autofill_dataset_converter.py
+++ b/chrome/test/functional/autofill_dataset_converter.py
+#!/usr/bin/python
+# Copyright (c) 2011 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""Converts profile datasets to dictionary list for Autofill profiles.
+
+Used for test autofill.AutoFillTest.testMergeDuplicateProfilesInAutofill.
+"""
+
+import codecs
+import logging
+import os
+import re
+import sys
+
+
+class DatasetConverter(object):
+  _fields = [
+    u'NAME_FIRST',
+    u'NAME_MIDDLE',
+    u'NAME_LAST',
+    u'EMAIL_ADDRESS',
+    u'COMPANY_NAME',
+    u'ADDRESS_HOME_LINE1',
+    u'ADDRESS_HOME_LINE2',
+    u'ADDRESS_HOME_CITY',
+    u'ADDRESS_HOME_STATE',
+    u'ADDRESS_HOME_ZIP',
+    u'ADDRESS_HOME_COUNTRY',
+    u'PHONE_HOME_WHOLE_NUMBER',
+    u'PHONE_FAX_WHOLE_NUMBER',
+  ]
+  _record_length = len(_fields)
+  _output_pattern = u'{'
+  for key in _fields:
+    _output_pattern += u"u'%s': u'%%s', " % key
+  _output_pattern = _output_pattern[:-1] + '},'
+  _re_single_quote = re.compile("'", re.UNICODE)
+  _logger = logging.getLogger(__name__)
+
+  def __init__(self, input_filename, output_filename=None,
+               logging_level=logging.ERROR):
+    """Constructs a dataset converter object.
+
+    Full input pattern:
+      '(?P<NAME_FIRST>.*?)\|(?P<MIDDLE_NAME>.*?)\|(?P<NAME_LAST>.*?)\|
+      (?P<EMAIL_ADDRESS>.*?)\|(?P<COMPANY_NAME>.*?)\|(?P<ADDRESS_HOME_LINE1>.*?)
+      \|(?P<ADDRESS_HOME_LINE2>.*?)\|(?P<ADDRESS_HOME_CITY>.*?)\|
+      (?P<ADDRESS_HOME_STATE>.*?)\|(?P<ADDRESS_HOME_ZIP>.*?)\|
+      (?P<ADDRESS_HOME_COUNTRY>.*?)\|
+      (?P<PHONE_HOME_WHOLE_NUMBER>.*?)\|(?P<PHONE_FAX_WHOLE_NUMBER>.*?)$'
+
+    Full ouput pattern:
+      "{u'NAME_FIRST': u'%s', u'NAME_MIDDLE': u'%s', u'NAME_LAST': u'%s',
+      u'EMAIL_ADDRESS': u'%s', u'COMPANY_NAME': u'%s', u'ADDRESS_HOME_LINE1':
+      u'%s', u'ADDRESS_HOME_LINE2': u'%s', u'ADDRESS_HOME_CITY': u'%s',
+      u'ADDRESS_HOME_STATE': u'%s', u'ADDRESS_HOME_ZIP': u'%s',
+      u'ADDRESS_HOME_COUNTRY': u'%s', u'PHONE_HOME_WHOLE_NUMBER': u'%s',
+      u'PHONE_FAX_WHOLE_NUMBER': u'%s',},"
+
+    Args:
+      input_filename: name and path of the input dataset.
+      output_filename: name and path of the converted file, default is none.
+      logging_level: set verbosity levels, default is ERROR.
+
+    Raises:
+      IOError: error if input file does not exist.
+    """
+    console = logging.StreamHandler()
+    console.setLevel(logging_level)
+    self._logger.addHandler(console)
+
+    self._input_filename = os.path.join(os.path.dirname(sys.argv[0]),
+                                        input_filename)
+    if not os.path.isfile(self._input_filename):
+      msg = 'File "%s" does not exist' % self._input_filename
+      self._logger.error(msg)
+      raise IOError(msg)
+    self._output_filename = output_filename
+
+  def _CreateDictionaryFromRecord(self, record):
+    """Constructs and returns a dictionary from a record in the dataset file.
+
+    Escapes single quotation first and uses split('|') to separate values.
+    The method assumes a valid record always contains at least one "|"
+    character.
+    Example:
+      Take an argument as a string u'John|Doe|Mountain View'
+      and returns a dictionary
+      {
+      u'NAME_FIRST': u'John',
+      u'NAME_LAST': u'Doe',
+      u'ADDRESS_HOME_CITY': u'Mountain View',
+      }
+
+    Args:
+      record: row of record from the dataset file.
+
+    Returns:
+      None if the current record line is invalid or a dictionary representing a
+      single record from the dataset file.
+    """
+    # Ignore irrelevant record lines that do not contain '|'.
+    if not '|' in record:
+      return
+    # Escaping single quote: "'" -> "\'"
+    record = self._re_single_quote.sub(r"\'", record)
+    record_list = record.split('|')
+    if record_list:
+      # Check for case when a record may have more or less fields than expected.
+      if len(record_list) != self._record_length:
+        self._logger.warning(
+            'A "|" separated line has %d fields instead of %d: %s' % (
+                len(record_list), self._record_length, record))
+        return
+      out_record = {}
+      for i, key in enumerate(self._fields):
+        out_record[key] = record_list[i]
+      return out_record
+
+  def Convert(self):
+    """Function to convert input data into the desired output format.
+
+    Returns:
+      List that holds all the dictionaries.
+    """
+    with open(self._input_filename) as input_file:
+      if self._output_filename:
+        output_file = codecs.open(self._output_filename, mode='wb',
+                                  encoding='utf-8-sig')
+      else:
+        output_file = None
+      try:
+        list_of_dict = []
+        i = 0
+        if output_file:
+          output_file.write('[')
+          output_file.write(os.linesep)
+        for line in input_file.readlines():
+          line = line.strip()
+          if not line:
+            continue
+          line = unicode(line, 'UTF-8')
+          output_record = self._CreateDictionaryFromRecord(line)
+          if output_record:
+            i += 1
+            list_of_dict.append(output_record)
+            output_line = self._output_pattern % tuple(
+                [output_record[key] for key in self._fields])
+            if output_file:
+              output_file.write(output_line)
+              output_file.write(os.linesep)
+            self._logger.info('%d: %s' % (i, line.encode(sys.stdout.encoding,
+                                                         'ignore')))
+            self._logger.info('\tconverted to: %s' %
+                              output_line.encode(sys.stdout.encoding, 'ignore'))
+        if output_file:
+          output_file.write(']')
+          output_file.write(os.linesep)
+        self._logger.info('%d lines converted SUCCESSFULLY!' % i)
+        self._logger.info('--- FINISHED ---')
+        return list_of_dict
+      finally:
+        if output_file:
+          output_file.close()
+
+
+def main():
+  c = DatasetConverter(r'../data/autofill/dataset.txt',
+                       r'../data/autofill/dataset_duplicate-profiles.txt',
+                       logging.INFO)
+  c.Convert()
+
+if __name__ == '__main__':
+  main()