Question

我正在创建一个表，并希望导入（加载到postgres）一个制表符分隔文件，该文件对某些列有一些缺失值（空单元格）。我希望这些单元格保持为空（或者可能在执行加载时将它们转换为NULL，或者在它们之后将它们保留为空）。

我尝试过以下脚本：

CREATE TABLE Conclusive_wg_H3K9me3_dn
(
Conclusive_wg_K9me3_DN_sorter float,    
Conclusive_wg_K9me3_DN_geneid character(80) NOT NULL,   
Conclusive_wg_K9me3_DN_bvi_srt float,   
Conclusive_wg_K9me3_DN_cbk_srt float,   
Conclusive_wg_K9me3_DN_dj_srt float,    
Conclusive_wg_K9me3_DN_evj_srt float,   
Conclusive_wg_K9me3_DN_flv_srt float,   
Conclusive_wg_K9me3_DN_ghw_srt float,   
Conclusive_wg_K9me3_DN_gvz_srt float,   
Conclusive_wg_K9me3_DN_srr_srt float,   
Conclusive_wg_K9me3_DN_AllCount float,  
Conclusive_wg_K9me3_DN_PercentAllCount float,   
Conclusive_wg_K9me3_DN_DynamicCounat float,     
Conclusive_wg_K9me3_DN_PercentDynamic float,    
Conclusive_wg_K9me3_DN_GeneName character(80) NOT NULL,
Conclusive_wg_K9me3_DN_RankSorting float,   
Conclusive_wg_K9me3_DN_NewScore float,
CONSTRAINT Conclusive_wg_K9me3_DN_geneid PRIMARY KEY (Conclusive_wg_K9me3_DN_geneid)
);
COPY Conclusive_wg_H3K9me3_dn FROM 'G:\CarrollLab\Teena\ConsolidationWithMethylationData_NoiseBelow32Removed\FromPostgreSQLdb\FigsAndTangentWholeGenes\DynamicWholeGeneValues\OutputFromPostgres\ConclusiveDynamicGenes\Conclusive_wg_H3K9me3_dn.txt' WITH (FORMAT 'csv', DELIMITER E'\t', NULL 'NULL',HEADER);

如果我使用文本分隔文件执行此脚本，其中所有行都具有所有浮点值，则导入过程可以正常工作！没有麻烦。

但是，在尝试使用带有几个空单元格的文件（缺少一些浮点值）运行它时，我遇到以下错误信息：

ERROR:  invalid input syntax for type double precision: ""
CONTEXT:  COPY conclusive_wg_h3k9me3_dn, line 9, column conclusive_wg_k9me3_dn_flv_srt: ""
********** Error **********

ERROR: invalid input syntax for type double precision: ""
SQL state: 22P02
Context: COPY conclusive_wg_h3k9me3_dn, line 9, column conclusive_wg_k9me3_dn_flv_srt: ""

实际上，我在txt分隔文件中的第9行在其中一列中缺少数据（空单元格）。

我是否必须首先在EXCEL（例如）中使用字符串＆＃34; NULL＆＃34;替换所有空表单元格。？... float类型会识别NULL字符串吗？

或者我可以让postgres以某种方式理解如何处理空单元格而不用其他东西替换空值？

如何解决这个问题？谢谢！罗伊

Answer 1

好的，问题解决了！

我应该将我的NULL分类更改为＆＃39; NULL＆＃39;相应地:(也见底部）： WITH（FORMAT＆＃39; csv＆＃39;，DELIMITER E＆＃39; \ t＆＃39;，NULL＆＃39;＆＃39;，HEADER）;

CREATE TABLE Conclusive_wg_H3K9me3_dn
(
Conclusive_wg_K9me3_DN_sorter float,    
Conclusive_wg_K9me3_DN_geneid character(80) NOT NULL,   
Conclusive_wg_K9me3_DN_bvi_srt float,   
Conclusive_wg_K9me3_DN_cbk_srt float,   
Conclusive_wg_K9me3_DN_dj_srt float,    
Conclusive_wg_K9me3_DN_evj_srt float,   
Conclusive_wg_K9me3_DN_flv_srt float,   
Conclusive_wg_K9me3_DN_ghw_srt float,   
Conclusive_wg_K9me3_DN_gvz_srt float,   
Conclusive_wg_K9me3_DN_srr_srt float,   
Conclusive_wg_K9me3_DN_AllCount float,  
Conclusive_wg_K9me3_DN_PercentAllCount float,   
Conclusive_wg_K9me3_DN_DynamicCounat float,     
Conclusive_wg_K9me3_DN_PercentDynamic float,    
Conclusive_wg_K9me3_DN_GeneName character(80) NOT NULL,
Conclusive_wg_K9me3_DN_RankSorting float,   
Conclusive_wg_K9me3_DN_NewScore float,
CONSTRAINT Conclusive_wg_K9me3_DN_geneid PRIMARY KEY (Conclusive_wg_K9me3_DN_geneid)
);
COPY Conclusive_wg_H3K9me3_dn FROM 'G:\CarrollLab\Teena\ConsolidationWithMethylationData_NoiseBelow32Removed\FromPostgreSQLdb\FigsAndTangentWholeGenes\DynamicWholeGeneValues\OutputFromPostgres\ConclusiveDynamicGenes\Conclusive_wg_H3K9me3_dn.txt' WITH (FORMAT 'csv', DELIMITER E'\t', NULL '',HEADER);

导入错过某些Float类型值的csv数据表

1 个答案: