ÐÏࡱáing to their origins. The groups are identified as the G-, T-, J-, K- and V-sources. For the purposes of ISO/IEC 10646-1 a unification process is applied to the ideographic characters taken from the codes in the source groups. In this process single ideographs from two or more of the source groups are associated together, and a single code position is assigned to them in this standard. The associations are made according to a set of procedures that are described below. Ideographs that are thus associated are described here as “unified”. NOTE - The unification process does not apply to the following collections of ideographic characters in the Basic multilingual Plane: - CJK RADICALS SUPPLEMENT (2E80 - 2EFF) - KANGXI RADICALS (2F00 - 2FDF) - CJK COMPATIBILITY IDEOGRAPHS (F900 - FAFF with the exception of FA1F and FA23). R.1. Unification procedure R.1.1 Scope of unification Ideographs that are unrelated in historical derivation (non-cognate characters) have not been unified. Example:  NOTE - The difference of shape between the two ideographs in the above example is in the length of the lower horizontal line. This is considered an actual difference of shape. Furthermore these ideographs have different meanings. The meaning of the first is "Soldier" and of the second is "Soil or Earth". An association between ideographs from different sources is made here if their shapes are sufficiently similar, according to the following system of classification. R.1.2 Two level classification A two-level system of classification is used to differentiate (a) between abstract shapes and (b) between actual shapes determined by particular typefaces. Variant forms of an ideograph, which can not be unified, are identified based on the difference between their abstract shapes. R.1.3 Procedure A unification procedure is used to determine whether two ideographs have the same abstract shape or different ones. The unification procedure has two stages, applied in the following order: a) Analysis of component structure; b) Analysis of component features; R.1.3.1 Analysis of component structure In the first stage of the procedure the component structure of each ideograph is examined. A component of an ideograph is a geometrical combination of primitive elements. Alternative ideographs can be configured from the same set of components. Components can be combined to create a new component with a more complicated structure. An ideograph, therefore, can be defined as a component tree, where the top node is the ideograph itself, and the bottom nodes are the primitive elements. This is shown in Figure R.1.  Figure R.1 - Component structure R.1.3.2 Analysis of component features In the second stage of the procedure, the components located at corresponding nodes of two ideographs are compared, starting from the most superior node, as shown in Figure R.2.  Figure R.2 - The most superior node of a component The following features of each ideograph to be compared are examined: a : the number of components, b : the relative position of the components in each complete ideograph, c : the structure of corresponding components. If one or more of the features (a to c above) are different between the ideographs in the comparison, the ideographs are considered to have different abstract shapes and are therefore not unified. If all of the features (a to c above) are the same between the ideographs, the ideographs are considered to have the same abstract shape and are therefore unified. R.1.4 Examples of differences of abstract shapes To illustrate rules derived from a: to c: in R.1.3.2, some typical examples of ideographs that are not unified, owing to differences of abstract shapes, are shown below. R.1.4.1 Different number of components The examples below illustrate rule a: since the two ideographs in each pair have different numbers of components. , ,  R.1.4.2 Different relative positions of components The examples below illustrate rule b:. Although the two ideographs in each pair have the same number of components, the relative positions of the components are different. ,  R.1.4.3 Different structure of a corresponding component The examples below illustrate rule c:. The structure of one (or more) corresponding components within the two ideographs in each pair is different. , , , , , , , , , , , , , , , , , , , ,  R.1.5 Differences of actual shapes To illustrate the classification described in R.1.2, some typical examples of ideographs that are unified are shown below. The two or three ideographs in each group below have different actual shapes, but they are considered to have the same abstract shape, and are therefore unified. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,  The differences are further classified according to the following examples. a) Differences in rotated strokes/dots , , , , ,  b) Differences in overshoot at the stroke initiation and/or termination , , , , , ,  c) Differences in contact of strokes , , , ,  d) Differences in protrusion at the folded corner of strokes  e) Differences in bent strokes  f) Differences in folding back at the stroke termination  g) Differences in accent at the stroke initiation , ,  h) Differences in "rooftop" modification ,  j) Combinations of the above differences  These differences in actual shapes of a unified ideograph are presented in the corresponding source columns for each code position entry in the code table in clause 27 of this International Standard. R.1.6 Source separation rule To preserve data integrity through multiple stages of code conversion (commonly known as “round-trip integrity”), any ideographs that are separately encoded in any one of the source standards listed below have not been unified. G-source: GB2312-80, GB12345-90, GB7589-87*, GB7590-87*, GB8565-88*, General Purpose Hanzi List for Modern Chinese Language* T-source: TCA-CNS 11643-1986/1st plane, TCA-CNS 11643-1986/2nd plane, TCA-CNS 11643-1986/14th plane* J-source: JIS X 0208-1990, JIS X 0212-1990 K-source: KS C 5601-1989, KS C 5657-1991 (A " * " after the reference number of a standard indicates that some of the ideographs included in that standard are not introduced into the unified collection.) However, some ideographs encoded in two standards belonging to the same source group ( e.g. GB2312-80 and GB12345-90 ) have been unified during the process of collecting ideographs from the source group. R.2. Arrangement procedure R.2.1 Scope of arrangement The arrangement of the CJK UNIFIED IDEOGRAPHS in the code table of clause 27 of this International Standard is based on the filing order of ideographs in the following dictionaries. Priority Dictionary Edition 1 Kangxi Dictionary  EMBED Word.Picture.6  Beijing 7th edition 2 Daikanwa Jiten  9th edition 3 Hanyu Dazidian  1st edition 4 Daejaweon  1st edition The dictionaries are used according to the priority order given in the table above. Priority 1 is highest. If an ideograph is found in one dictionary, the dictionaries of lower priority are not examined. R.2.2 Procedure R.2.2.1 Ideographs found in the dictionaries a) If an ideograph is found in the Kangxi Dictionary, it is positioned in the code table in accordance with the Kangxi Dictionary order. b) If an ideograph is not found in the Kangxi Dictionary but is found in the Daikanwa Jiten, it is given a position at the end of the radical-stroke group under which is indexed the nearest preceding Daikanwa Jiten character that also appears in the Kangxi dictionary. c) If an ideograph is found in neither the Kangxi nor the Daikanwa, the Hanyu Dazidian and the Daejaweon dictionaries are referred to with a similar procedure. R.2.2.2 Ideographs not found in the dictionaries If an ideograph is not found in any of the four dictionaries, it is given a position at the end of the radical-stroke group (after the characters that are present in the dictionaries) and it is indexed under the same radical-stroke count. R.3. Source code separation examples The pairs (or triplets) of ideographs shown below are exceptions to the unification rules described in clause R.1 of this Annex. They are not unified because of the source code separation rule described in clause R.1.6. NOTES 1. The particular source code group (or groups) that causes the source code separation rule to apply is indicated by the letter (G, J, K, or T) that appears to the right of each pair (or triplet) of ideographs. The source code groups that correspond to these letters are identified at the beginning of this Annex. 2. The ideograph pairs are listed below in ascending order by the code position of the first ideograph of each pair. The sequence progresses downwards in the left column as far as each marker ( v------v ), and then continues downwards in the adjoining right column, starting at the previous marker.  T  T 4E1F 4E22 5156 5157  GT  TJ 4E48 5E7A 518A 518C  GTJ  G 4E89 722D 51C0 51C8  J  T 4EDE 4EED 51E2 51E3  T  TJ 4F75 5002 5203 5204  T  TJ 4FA3 4FB6 520A 520B  TJK  T 4FC1 4FE3 5220 522A  T  T 4FDE 516A 5225 522B  T  TJ 4FF1 5036 5238 52B5  T  T 5024 503C 5239 524E  T  T 5077 5078 524F 5259  TJ  T 507D 50DE 525D 5265  T  J 514C 5151 5292 5294  TJ  T 514E 5154 52FB 5300 V------------------------------V  T  T 5355 5358 5848 588D  TK  TJ 5373 537D 5861 586B  TJ  T 5377 5DFB 5897 589E  GT  GTJ 53C1 53C2 58EE 58EF  T  T 53C3 53C4 58FD 5900  T  T 5415 5442 5910 657B  T  GTJ 541E 5451 5932 672C  TJ  J 5433 5434 5449 5965 5967  T  TJ 5436 5450 5968 596C 734E  T  GT 543F 544A 5986 599D  T  T 5527 559E 598D 59F8  T  T 55A9 55BB 59CD 59D7  T  GT 5618 5653 59EB 59EC  GTJ  T 568F 5694 5A1B 5A2F 5A31  T  T 56EF 56FD 5A55 5AAB  TJ  T 5708 570F 5A7E 5AAE  T  TK 570E 5713 5AAA 5ABC  T  T 5716 5717 5AAF 5B00  T  T 5759 5DE0 5B0E 5B14  J  GT 57D2 57D3 5B24 5B37 V------------------------------V  T  T 5B73 5B76 5F39 5F3E  T  TJ 5BAB 5BAE 5F50 5F51  T  T 5BDB 5BEC 5F54 5F55  T  T 5BDC 5BE7 5F59 5F5A  GTJ  J 5BDD 5BE2 5F5B 5F5C  J  T 5C02 5C08 5F5D 5F5E  GTJ  T 5C06 5C07 5F65 5F66  T  T 5C13 5C14 5FB3 5FB7  T  T 5C19 5C1A 5FB4 5FB5  T  TJ 5C2A 5C2B 6075 60E0  T  T 5C36 5C37 6085 60A6  T  T 5C4F 5C5B 609E 60AE  GT  T 5CE5 5D22 60B3 60EA  T  T 5DD3 5DD4 6120 614D  T  TJ 5E21 5E32 613C 614E  TJ  GT 5E2F 5E36 6229 622C  T  T 5E76 5E77 622F 6231  T  T 5EC4 5ECF 6236 6237 6238  T  T 5F11 5F12 623B 623E  T  T 5F37 5F3A 629B 62CB V------------------------------V  TJ  T 629C 62D4 6961 6986  T  T 6329 635D 6982 69EA  TJ  T 633F 63D2 63F7 6985 69B2  TJ  T 634F 63D1 699D 6A27  TJ  J 635C 641C 69C7 69D9  T  TJ 63B2 63ED 69D8 6A23  TJ  T 63FA 6416 6447 6A2A 6A6B  T  T 63FE 6435 6B65 6B69  TJ  T 6483 64CA 6B72 6B73  T  T 654E 6559 6B7F 6B81  T  GTJ 6553 655A 6BBB 6BBC  T  T 65E2 65E3 6BC0 6BC1  T  T 6602 663B 6BCE 6BCF  T  T 665A 6669 6C32 6C33  T  T 66A8 66C1 6C5A 6C61  J  TJ 66FD 66FE 6C92 6CA1  T TJ 67B4 67FA 6D44 6DE8  T  T 67E5 67FB 6D89 6E09  T  T 67F5 6805 6D97 6D9A  T  T 68B2 68C1 6D99 6DDA V------------------------------V  T  TJK 6DE5 6E0C 773E 8846  T  T 6DF8 6E05 7814 784F  T  TJ 6E07 6E34 797F 7984  T  T 6E29 6EAB 79BF 79C3  T  T 6E88 6F59 7A05 7A0E  T  TJ 6E89 6F11 7A42 7A57  T  GJ 6EDA 6EFE 7B5D 7B8F  GTJK  T 6F5B 6FF3 7BB3 7C08  T  T 7028 702C 7BE1 7C12  GTJ  T 70BA 7232 7CA4 7CB5  GTJK  T 712D 7162 7D55 7D76  J  T 7155 7199 7DA0 7DD1  T  T 7174 7185 7DD2 7DD6  GT  T 72B6 72C0 7DE3 7E01  TJ  T 7464 7476 7DFC 7E15  T  T 74F6 7501 7E48 7E66  T  TJ 7522 7523 7FAE 7FB9  J  T 75E9 7626 7FF6 7FFA  T  T 76A1 76A5 80FC 8141  TJ  T 771E 771F 812B 8131 V------------------------------V  T  J 817D 8183 8B20 8B21  GT  T 8203 8204 8C5C 8C63  TJ  TJ 820D 820E 8D70 8D71  J  T 8216 8217 8EFF 8F27  TJ  J 8358 838A 8F1C 8F3A  TJ  T 83D1 8458 8F3C 8F40  T  T 8480 8495 8FBE 8FD6  GJ  TJ 848B 8523 8FF8 902C  T  J 848D 853F 9059 9065  T  T 8570 8580 90A2 90C9  T  T 85AB 85B0 90CE 90DE  T  T 85F4 860A 90F7 9109 9115  T  T 865A 865B 9196 919E  T  J 86FB 8715 91A4 91AC  TJK  T 885B 885E 9203 9292  TK  T 886E 889E 92B3 92ED  GJK  T 88C5 88DD 9304 9332  T  TK 8A2E 8A7D 932C 934A  T  TJ 8AAA 8AAC 93AD 93AE  TJ  T 8ACC 8AEB 95B1 95B2 V------------------------------V  G  T 9667 9689 9AD8 9AD9  T  TJ 9751 9752 9AEA 9AEE  GTJ  T 9759 975C 9B2C 9B2D  J  TJ 976D 9771 9C1B 9C2E  T  T 9839 983D 9CEF 9CF3  TJ  J 984F 9854 9D87 9DAB  J  J 985A 985B 9DC6 9DCF  J  T 98EE 98F2 9EAA 9EAB  TJ  T 9905 9920 9EBC 9EBD  TJK  T 99B1 99C4 9EC3 9EC4  TK  T 99E2 9A08 9ED1 9ED2  T 9AA9 9AAB In accordance with the unification procedures described in R.1 of this Annex the pairs (or triplets) of ideographs shown below are not unified. The reason for non-unification is indicated by the reference which appears to the right of each pair (or triplet). For “non-cognate” see R.1.1 NOTE - The reason for non-unification in these examples is different from the source code separation rule described in clause R.1.6. non cognate  non cognate 5191 80C4 6710 80CA  R.1.4.3  non cognate 51B2 6C96 6713 8101  R.1.4.3  non cognate 51B3 6C7A 6718 8127  R.1.4.3  non cognate 51B5 6CC1 6723 81A7  R.1.4.3  R.1.4.3 579B 579C 6735 6736  R.1.4.2  R.1.4.3 5B7C 5B7D 7054 7067  R.1.4.3  R.1.4.3 5BF3 5BF6 7A32 7A3B  R.1.4.1  R.1.4.3 5EF0 5EF3 7FF1 7FF6  R.1.4.1  R.1.4.3 61D0 61F7 8007 8008 8009  R.1.4.3  R.1.4.1 6560 656A 8074 807C 807D non cognate  R.1.4.2 670C 80A6 8346 834A non cognate  R.1.4.3 670F 80D0 8EB1 8EB2 Draft 2 for ISO/IEC 10646-1:1999(E) ( ISO/IEC ( ISO/IEC Draft 2 for ISO/IEC 10646-1:1999(E) PAGE 70 PAGE 77 ISO/IEC 10646-1:1993/DAM.8 most superior node