diff --git a/main1.py b/main1.py index 697b2a8..5d72d37 100644 --- a/main1.py +++ b/main1.py @@ -1,21 +1,132 @@ -import xml.etree.ElementTree as ET -from bs4 import BeautifulSoup +from lxml import etree -def open_file(file_path): - - - with open(file_path, 'r') as f: - data = f.read() - Bs_data = BeautifulSoup(data, "xml") - b_unique = Bs_data.find_all('unique') +data=[] +class Data_farame(): + def __init__(self,section_KPTR,cadastral_number,cadastral_quarter,object_type,land_plot_area,error_in_determining_area,availability_coordinates,explanations): + self.section_KPTR = section_KPTR #Раздел КПТР + self.cadastral_number = cadastral_number #Кадастровый номер + self.cadastral_quarter = cadastral_quarter #Кадастровый квартал + self.object_type = object_type #Вид объекта + self.land_plot_area = land_plot_area #Площадь ЗУ + self.error_in_determining_area = error_in_determining_area #Погрешность определения площади + self.availability_coordinates = availability_coordinates #наличие координат + self.explanations = explanations #Пояснения + +class SpecifyParcel(): - print(b_unique) - b_name = Bs_data.find('child', {'name':'Frank'}) + def chec_koordinat(new_parcel): + if 'EntitySpatial' in new_parcel: + if 'SpatialElement' in new_parcel['EntitySpatial']: + return "да" + else: + return "нет" + else: + return "нет" - print(b_name) - value = b_name.get('test') + def chec_tag_ProvidingPassCadastralNumbers(ProvidingPassCadastralNumbers): + if 'ProvidingPassCadastralNumbers' in ProvidingPassCadastralNumbers['ExistParcel'][0]: + return ProvidingPassCadastralNumbers['ExistParcel'][0]['ProvidingPassCadastralNumbers'][0]['Other'][0]['#text'] + else: + return "" + + def chec_tag_Explanation(Explanation): + if 'Explanation' in Explanation['ExistParcel'][0]: + return Explanation['ExistParcel'][0]['Explanation'][0]['#text'] + else: + return "" - print(value) + def __init__(self): + try: + for i,item in enumerate(xml_dict['Package'][0]['SpecifyParcels'][0]['SpecifyParcel']): + data.append( + Data_farame( + section_KPTR="Уточнение ЗУ", + cadastral_number=item['ExistParcel'][0]['@attributes']["CadastralNumber"], + cadastral_quarter=item['ExistParcel'][0]['CadastralBlock'][0]['#text'], + object_type=self.chec_tag_ProvidingPassCadastralNumbers(ProvidingPassCadastralNumbers=item), + land_plot_area=item['ExistParcel'][0]['Area'][0]['Area'][0]['#text'], + error_in_determining_area=item['ExistParcel'][0]['Area'][0]['Inaccuracy'][0]['#text'], + availability_coordinates=self.chec_koordinat(item['ExistParcel'][0]['EntitySpatial'][0]), + explanations=self.chec_tag_Explanation(Explanation=item) + ) + ) + except: + return + +class NewParcel: -open_file("C:/Users/jze9/Documents/work/git/kadastr12/test/MapPlanTerritory_E2DF5B75-E136-48DA-BB4C-8887BA4ABF46.xml") \ No newline at end of file + def add_new_data(self,doc_category): + data="" + for item in doc_category: + data += f"{item}: {doc_category[item][0]['#text']}, " + return data + + def chec_koordinat(self,new_parcel): + if 'EntitySpatial' in new_parcel: + if 'SpatialElement' in new_parcel['EntitySpatial'][0]: + return "да" + else: + return "нет" + else: + return "нет" + + def get_cad_number(self,object_realty): + cad_nimbers = "" + for item in object_realty: + cad_nimbers+=f"{item['InnerCadastralNumbers'][0]['CadastralNumber'][0]['#text']}, " + return cad_nimbers + + def __init__(self,xml_dict): + try: + for new_parcel in xml_dict['Package'][0]['FormParcels'][0]['NewParcel']: + data.append( + Data_farame( + section_KPTR="Образование ЗУ", + cadastral_number=self.get_cad_number(new_parcel['ObjectsRealty'][0]['ObjectRealty']), + cadastral_quarter=new_parcel['CadastralBlock'][0]['#text'], + object_type="", + land_plot_area=new_parcel['Area'][0]['Area'][0]['#text'], + error_in_determining_area=new_parcel['Area'][0]['Inaccuracy'][0]['#text'], + availability_coordinates=self.chec_koordinat(new_parcel), + explanations=self.add_new_data(new_parcel['Category'][0]['DocCategory'][0]) + ) + ) + except: + return + +def lxml_to_dict(element): + result = {} + + # Обрабатываем атрибуты (если есть) + if element.attrib: + result["@attributes"] = element.attrib + + # Обрабатываем текстовое содержимое (если есть) + if element.text and element.text.strip(): + result["#text"] = element.text.strip() + + # Группируем все дочерние элементы в списки + children_by_tag = {} + for child in element: + child_tag = child.tag + if child_tag not in children_by_tag: + children_by_tag[child_tag] = [] + children_by_tag[child_tag].append(lxml_to_dict(child)) + + # Добавляем в результат (всегда списками) + for tag, children in children_by_tag.items(): + result[tag] = children # Всегда массив, даже если элемент один + + return result + +#tree = etree.parse('C:/Users/USER/Downloads/12. ХМЛ КПТР в ексель/Исходные данные/Карта-план в хмл прмиер Копейск на кк 74 30 0104002/MapPlanTerritory_54F3719F-2ED8-48FF-8442-9813191EBA51.xml') +tree = etree.parse('C:/Users/USER/Downloads/12. ХМЛ КПТР в ексель/Исходные данные/Карта-план в хмл прмиер Копейск на кк 74 30 0104002/MapPlanTerritory_E2DF5B75-E136-48DA-BB4C-8887BA4ABF46.xml') +xml_dict = lxml_to_dict(tree.getroot()) +temp = NewParcel(xml_dict) + + + + +pass + diff --git a/pyproject.toml b/pyproject.toml index b55ad02..9b0ce7d 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,4 +6,5 @@ readme = "README.md" requires-python = ">=3.12" dependencies = [ "bs4>=0.0.2", + "lxml>=5.4.0", ] diff --git a/uv.lock b/uv.lock index 0336edc..f1574ec 100644 --- a/uv.lock +++ b/uv.lock @@ -27,6 +27,48 @@ wheels = [ { url = "https://files.pythonhosted.org/packages/51/bb/bf7aab772a159614954d84aa832c129624ba6c32faa559dfb200a534e50b/bs4-0.0.2-py2.py3-none-any.whl", hash = "sha256:abf8742c0805ef7f662dce4b51cca104cffe52b835238afc169142ab9b3fbccc", size = 1189, upload-time = "2024-01-17T18:15:48.613Z" }, ] +[[package]] +name = "lxml" +version = "5.4.0" +source = { registry = "https://pypi.org/simple" } +sdist = { url = "https://files.pythonhosted.org/packages/76/3d/14e82fc7c8fb1b7761f7e748fd47e2ec8276d137b6acfe5a4bb73853e08f/lxml-5.4.0.tar.gz", hash = "sha256:d12832e1dbea4be280b22fd0ea7c9b87f0d8fc51ba06e92dc62d52f804f78ebd", size = 3679479, upload-time = "2025-04-23T01:50:29.322Z" } +wheels = [ + { url = "https://files.pythonhosted.org/packages/f8/4c/d101ace719ca6a4ec043eb516fcfcb1b396a9fccc4fcd9ef593df34ba0d5/lxml-5.4.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:b5aff6f3e818e6bdbbb38e5967520f174b18f539c2b9de867b1e7fde6f8d95a4", size = 8127392, upload-time = "2025-04-23T01:46:04.09Z" }, + { url = "https://files.pythonhosted.org/packages/11/84/beddae0cec4dd9ddf46abf156f0af451c13019a0fa25d7445b655ba5ccb7/lxml-5.4.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:942a5d73f739ad7c452bf739a62a0f83e2578afd6b8e5406308731f4ce78b16d", size = 4415103, upload-time = "2025-04-23T01:46:07.227Z" }, + { url = "https://files.pythonhosted.org/packages/d0/25/d0d93a4e763f0462cccd2b8a665bf1e4343dd788c76dcfefa289d46a38a9/lxml-5.4.0-cp312-cp312-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:460508a4b07364d6abf53acaa0a90b6d370fafde5693ef37602566613a9b0779", size = 5024224, upload-time = "2025-04-23T01:46:10.237Z" }, + { url = "https://files.pythonhosted.org/packages/31/ce/1df18fb8f7946e7f3388af378b1f34fcf253b94b9feedb2cec5969da8012/lxml-5.4.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:529024ab3a505fed78fe3cc5ddc079464e709f6c892733e3f5842007cec8ac6e", size = 4769913, upload-time = "2025-04-23T01:46:12.757Z" }, + { url = "https://files.pythonhosted.org/packages/4e/62/f4a6c60ae7c40d43657f552f3045df05118636be1165b906d3423790447f/lxml-5.4.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7ca56ebc2c474e8f3d5761debfd9283b8b18c76c4fc0967b74aeafba1f5647f9", size = 5290441, upload-time = "2025-04-23T01:46:16.037Z" }, + { url = "https://files.pythonhosted.org/packages/9e/aa/04f00009e1e3a77838c7fc948f161b5d2d5de1136b2b81c712a263829ea4/lxml-5.4.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a81e1196f0a5b4167a8dafe3a66aa67c4addac1b22dc47947abd5d5c7a3f24b5", size = 4820165, upload-time = "2025-04-23T01:46:19.137Z" }, + { url = "https://files.pythonhosted.org/packages/c9/1f/e0b2f61fa2404bf0f1fdf1898377e5bd1b74cc9b2cf2c6ba8509b8f27990/lxml-5.4.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00b8686694423ddae324cf614e1b9659c2edb754de617703c3d29ff568448df5", size = 4932580, upload-time = "2025-04-23T01:46:21.963Z" }, + { url = "https://files.pythonhosted.org/packages/24/a2/8263f351b4ffe0ed3e32ea7b7830f845c795349034f912f490180d88a877/lxml-5.4.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:c5681160758d3f6ac5b4fea370495c48aac0989d6a0f01bb9a72ad8ef5ab75c4", size = 4759493, upload-time = "2025-04-23T01:46:24.316Z" }, + { url = "https://files.pythonhosted.org/packages/05/00/41db052f279995c0e35c79d0f0fc9f8122d5b5e9630139c592a0b58c71b4/lxml-5.4.0-cp312-cp312-manylinux_2_28_ppc64le.whl", hash = "sha256:2dc191e60425ad70e75a68c9fd90ab284df64d9cd410ba8d2b641c0c45bc006e", size = 5324679, upload-time = "2025-04-23T01:46:27.097Z" }, + { url = "https://files.pythonhosted.org/packages/1d/be/ee99e6314cdef4587617d3b3b745f9356d9b7dd12a9663c5f3b5734b64ba/lxml-5.4.0-cp312-cp312-manylinux_2_28_s390x.whl", hash = "sha256:67f779374c6b9753ae0a0195a892a1c234ce8416e4448fe1e9f34746482070a7", size = 4890691, upload-time = "2025-04-23T01:46:30.009Z" }, + { url = "https://files.pythonhosted.org/packages/ad/36/239820114bf1d71f38f12208b9c58dec033cbcf80101cde006b9bde5cffd/lxml-5.4.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:79d5bfa9c1b455336f52343130b2067164040604e41f6dc4d8313867ed540079", size = 4955075, upload-time = "2025-04-23T01:46:32.33Z" }, + { url = "https://files.pythonhosted.org/packages/d4/e1/1b795cc0b174efc9e13dbd078a9ff79a58728a033142bc6d70a1ee8fc34d/lxml-5.4.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3d3c30ba1c9b48c68489dc1829a6eede9873f52edca1dda900066542528d6b20", size = 4838680, upload-time = "2025-04-23T01:46:34.852Z" }, + { url = "https://files.pythonhosted.org/packages/72/48/3c198455ca108cec5ae3662ae8acd7fd99476812fd712bb17f1b39a0b589/lxml-5.4.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:1af80c6316ae68aded77e91cd9d80648f7dd40406cef73df841aa3c36f6907c8", size = 5391253, upload-time = "2025-04-23T01:46:37.608Z" }, + { url = "https://files.pythonhosted.org/packages/d6/10/5bf51858971c51ec96cfc13e800a9951f3fd501686f4c18d7d84fe2d6352/lxml-5.4.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:4d885698f5019abe0de3d352caf9466d5de2baded00a06ef3f1216c1a58ae78f", size = 5261651, upload-time = "2025-04-23T01:46:40.183Z" }, + { url = "https://files.pythonhosted.org/packages/2b/11/06710dd809205377da380546f91d2ac94bad9ff735a72b64ec029f706c85/lxml-5.4.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:aea53d51859b6c64e7c51d522c03cc2c48b9b5d6172126854cc7f01aa11f52bc", size = 5024315, upload-time = "2025-04-23T01:46:43.333Z" }, + { url = "https://files.pythonhosted.org/packages/f5/b0/15b6217834b5e3a59ebf7f53125e08e318030e8cc0d7310355e6edac98ef/lxml-5.4.0-cp312-cp312-win32.whl", hash = "sha256:d90b729fd2732df28130c064aac9bb8aff14ba20baa4aee7bd0795ff1187545f", size = 3486149, upload-time = "2025-04-23T01:46:45.684Z" }, + { url = "https://files.pythonhosted.org/packages/91/1e/05ddcb57ad2f3069101611bd5f5084157d90861a2ef460bf42f45cced944/lxml-5.4.0-cp312-cp312-win_amd64.whl", hash = "sha256:1dc4ca99e89c335a7ed47d38964abcb36c5910790f9bd106f2a8fa2ee0b909d2", size = 3817095, upload-time = "2025-04-23T01:46:48.521Z" }, + { url = "https://files.pythonhosted.org/packages/87/cb/2ba1e9dd953415f58548506fa5549a7f373ae55e80c61c9041b7fd09a38a/lxml-5.4.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:773e27b62920199c6197130632c18fb7ead3257fce1ffb7d286912e56ddb79e0", size = 8110086, upload-time = "2025-04-23T01:46:52.218Z" }, + { url = "https://files.pythonhosted.org/packages/b5/3e/6602a4dca3ae344e8609914d6ab22e52ce42e3e1638c10967568c5c1450d/lxml-5.4.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ce9c671845de9699904b1e9df95acfe8dfc183f2310f163cdaa91a3535af95de", size = 4404613, upload-time = "2025-04-23T01:46:55.281Z" }, + { url = "https://files.pythonhosted.org/packages/4c/72/bf00988477d3bb452bef9436e45aeea82bb40cdfb4684b83c967c53909c7/lxml-5.4.0-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9454b8d8200ec99a224df8854786262b1bd6461f4280064c807303c642c05e76", size = 5012008, upload-time = "2025-04-23T01:46:57.817Z" }, + { url = "https://files.pythonhosted.org/packages/92/1f/93e42d93e9e7a44b2d3354c462cd784dbaaf350f7976b5d7c3f85d68d1b1/lxml-5.4.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cccd007d5c95279e529c146d095f1d39ac05139de26c098166c4beb9374b0f4d", size = 4760915, upload-time = "2025-04-23T01:47:00.745Z" }, + { url = "https://files.pythonhosted.org/packages/45/0b/363009390d0b461cf9976a499e83b68f792e4c32ecef092f3f9ef9c4ba54/lxml-5.4.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0fce1294a0497edb034cb416ad3e77ecc89b313cff7adbee5334e4dc0d11f422", size = 5283890, upload-time = "2025-04-23T01:47:04.702Z" }, + { url = "https://files.pythonhosted.org/packages/19/dc/6056c332f9378ab476c88e301e6549a0454dbee8f0ae16847414f0eccb74/lxml-5.4.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24974f774f3a78ac12b95e3a20ef0931795ff04dbb16db81a90c37f589819551", size = 4812644, upload-time = "2025-04-23T01:47:07.833Z" }, + { url = "https://files.pythonhosted.org/packages/ee/8a/f8c66bbb23ecb9048a46a5ef9b495fd23f7543df642dabeebcb2eeb66592/lxml-5.4.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:497cab4d8254c2a90bf988f162ace2ddbfdd806fce3bda3f581b9d24c852e03c", size = 4921817, upload-time = "2025-04-23T01:47:10.317Z" }, + { url = "https://files.pythonhosted.org/packages/04/57/2e537083c3f381f83d05d9b176f0d838a9e8961f7ed8ddce3f0217179ce3/lxml-5.4.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:e794f698ae4c5084414efea0f5cc9f4ac562ec02d66e1484ff822ef97c2cadff", size = 4753916, upload-time = "2025-04-23T01:47:12.823Z" }, + { url = "https://files.pythonhosted.org/packages/d8/80/ea8c4072109a350848f1157ce83ccd9439601274035cd045ac31f47f3417/lxml-5.4.0-cp313-cp313-manylinux_2_28_ppc64le.whl", hash = "sha256:2c62891b1ea3094bb12097822b3d44b93fc6c325f2043c4d2736a8ff09e65f60", size = 5289274, upload-time = "2025-04-23T01:47:15.916Z" }, + { url = "https://files.pythonhosted.org/packages/b3/47/c4be287c48cdc304483457878a3f22999098b9a95f455e3c4bda7ec7fc72/lxml-5.4.0-cp313-cp313-manylinux_2_28_s390x.whl", hash = "sha256:142accb3e4d1edae4b392bd165a9abdee8a3c432a2cca193df995bc3886249c8", size = 4874757, upload-time = "2025-04-23T01:47:19.793Z" }, + { url = "https://files.pythonhosted.org/packages/2f/04/6ef935dc74e729932e39478e44d8cfe6a83550552eaa072b7c05f6f22488/lxml-5.4.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:1a42b3a19346e5601d1b8296ff6ef3d76038058f311902edd574461e9c036982", size = 4947028, upload-time = "2025-04-23T01:47:22.401Z" }, + { url = "https://files.pythonhosted.org/packages/cb/f9/c33fc8daa373ef8a7daddb53175289024512b6619bc9de36d77dca3df44b/lxml-5.4.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4291d3c409a17febf817259cb37bc62cb7eb398bcc95c1356947e2871911ae61", size = 4834487, upload-time = "2025-04-23T01:47:25.513Z" }, + { url = "https://files.pythonhosted.org/packages/8d/30/fc92bb595bcb878311e01b418b57d13900f84c2b94f6eca9e5073ea756e6/lxml-5.4.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:4f5322cf38fe0e21c2d73901abf68e6329dc02a4994e483adbcf92b568a09a54", size = 5381688, upload-time = "2025-04-23T01:47:28.454Z" }, + { url = "https://files.pythonhosted.org/packages/43/d1/3ba7bd978ce28bba8e3da2c2e9d5ae3f8f521ad3f0ca6ea4788d086ba00d/lxml-5.4.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:0be91891bdb06ebe65122aa6bf3fc94489960cf7e03033c6f83a90863b23c58b", size = 5242043, upload-time = "2025-04-23T01:47:31.208Z" }, + { url = "https://files.pythonhosted.org/packages/ee/cd/95fa2201041a610c4d08ddaf31d43b98ecc4b1d74b1e7245b1abdab443cb/lxml-5.4.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:15a665ad90054a3d4f397bc40f73948d48e36e4c09f9bcffc7d90c87410e478a", size = 5021569, upload-time = "2025-04-23T01:47:33.805Z" }, + { url = "https://files.pythonhosted.org/packages/2d/a6/31da006fead660b9512d08d23d31e93ad3477dd47cc42e3285f143443176/lxml-5.4.0-cp313-cp313-win32.whl", hash = "sha256:d5663bc1b471c79f5c833cffbc9b87d7bf13f87e055a5c86c363ccd2348d7e82", size = 3485270, upload-time = "2025-04-23T01:47:36.133Z" }, + { url = "https://files.pythonhosted.org/packages/fc/14/c115516c62a7d2499781d2d3d7215218c0731b2c940753bf9f9b7b73924d/lxml-5.4.0-cp313-cp313-win_amd64.whl", hash = "sha256:bcb7a1096b4b6b24ce1ac24d4942ad98f983cd3810f9711bcd0293f43a9d8b9f", size = 3814606, upload-time = "2025-04-23T01:47:39.028Z" }, +] + [[package]] name = "soupsieve" version = "2.7" @@ -42,10 +84,14 @@ version = "0.1.0" source = { virtual = "." } dependencies = [ { name = "bs4" }, + { name = "lxml" }, ] [package.metadata] -requires-dist = [{ name = "bs4", specifier = ">=0.0.2" }] +requires-dist = [ + { name = "bs4", specifier = ">=0.0.2" }, + { name = "lxml", specifier = ">=5.4.0" }, +] [[package]] name = "typing-extensions"